Abstract

Book
    

THE 2nd INTERNATIONAL CONFERENCE

OF ADVANCED COMPUTING AND
INFORMATICS (ICACIN 2021)

May 24-25, 2021

Casablanca, Morocco

Advances on Smart Computing and Informatics

Editors:

Faisal Saeed
Tawfik Al-Hadhrami
Errais Mohammed
Mohammed AlSarem
    
                             Organizing  Committee
Honorary Chair
Saddiqi Omar Dean of the Faculty of science,
Ain Chock, Hassan II University
Casablanca, Morocco

International Advisory Board

Naomie Salim Universiti Teknologi Malaysia, Malaysia
Rose Alinda Alias Universiti Teknologi Malaysia, Malaysia
Ahmad Lotfi, Nottingham Trent University, UK
Funminiyi Olajide Nottingham Trent University, UK
Fahad Ghabban Taibah University, Kingdom of Saudi Arabia
Muhannad Al-Muhaimeed Taibah University, Kingdom of Saudi Arabia
Mohammed Alshargabi Najran University, Kingdom of Saudi Arabia

Conference General Chair

Errais Mohammed Hassan II University, Casablanca, Morocco
Faisal Saeed President, Yemeni Scientists Research Group
(YSRG)
Head of Data Science Research Group in
Taibah University, Kingdom of Saudi Arabia

Program Committee Co-Chairs

Mohamed Lahby Hassan II University - Casablanca, Morocco
Raouyane Brahim, Hassan II University – Casablanca Morocco
Mohammed Al-Sarem Taibah University, Kingdom of Saudi Arabia

Publications Committee
Tawfik Al-Hadhrami Nottingham Trent University, UK

ii
    
                             Organizing  Committee

Publicity Committee
Mandar Meriem National School of Applied Science of Berre-
chid – Morocco
IT Committee
Koulali Rim Hassan II University, Casablanca, Morocco
Chbihi Louhdi Mohammed Réda Hassan II University, Casablanca, Morocco

Secretary
Baddi youssef Chouayb Doukkali University, Morocco
Belmakki Mostapha INPT – Rabat, Morocco
Nadhmi Gazem Taibah University, Kingdom of Saudi Arabia

Finance Committee
Mohammed ERRAIS Hassan II University, Casablanca, Morocco
Abdullah Dahawi Universiti Teknologi Malaysia, Malaysia

Logistic Committee
Samia Benabdellah Hassan II University, Casablanca, Morocco
Dehbi Rachid Hassan II University, Casablanca, Morocco
Mohamed Rachdi Hassan II University, Casablanca, Morocco

Registration Committee
Raouyane Brahim Hassan II University, Casablanca, Morocco
Lahbi Mohammed Hassan II University, Casablanca, Morocco
Mohamed Rachdi Hassan II University, Casablanca, Morocco

iii
    
                             Organizing  Committee
Sponsorship Committee
Jai Andaloussi Said Hassan II University, Casablanca, Morocco
Rida Mohammed Hassan II University, Casablanca, Morocco
Fejtah Leila Hassan II University, Casablanca, Morocco
Ailane Abdellah, ENSAK, Morocco
Oukaira Aziz University of Quebec in Outaouais, Canada

Track Chairs
Abdelhamid Emara Taibah University, Saudi Arabia
Ali Ben Abbes University of Manouba, Tunisia
Lahbi Mohammed Hassan II University, Casablanca, Morocco
Maha Idriss Taibah University, Saudi Arabia
Nouf Alharbi Taibah University, Saudi Arabia
Wadii Boulila University of Manouba, Tunisia
Mandar Meriem Hassan II University, Casablanca Morocco
Youssef Baddi Chouhaib Doukkali University, El Jadida ,
Morocco

iv
  
Technical  Committee  

Aamre Khalil, Lorraine University, France

Abbas Dandache, Lorraine University, France
Abdallah AlSaeedi, Taibah University, KSA
Abdelaziz Ettaoufik, Hassan II University – Casablanca Morocco
Abdellah Kassem, UND, Lebanon
Abdelltif EL Byed, Hassan II University – Casablanca Morocco
Abderhmane Jarrou, Lorraine University, France
Abderrahim Maizat , Hassan II University – Casablanca Morocco
Abderrahim Tragha, Hassan II University – Casablanca Morocco
Abdullah Ghareb, University of Saba Region, Yemen
Abdulrahman Alsewari, Universiti Malaysia Pahang, Malaysia
Abdulrazak Alhababi, UNIMAS, Malaysia
Adil Sayouti, Ecole Royale Navale, Morocco
Ahmad Abu-Al-Aish , Jerash University, Jordan
Ahmad Al Shami, Higher Colleges of Technology, UAE
Ahmed Lakhssassi, University of Quebec in Outaouais, Canada
Ahmed Madian, NISC, Egypt
Ahlem Ferchichi, University of Manouba, Tunisia
Akrem Sellami, University of Lorraine, France
Ali Balaid, Aden University, Yemen
Ali Jwaid, Nottingham Trent University, UK
Ali Zahrawi, Yarmouk University, Jordan
Anmar Abuhamdah , Taibah University, Saudi Arabia
Ashraf Osman, Alzaiem Alazhari University, Sudan
Asmaa Benghabrit, ENIM, Rabat - Morocco
Assia Bakali , Ecole Royale Navale, Morocco
Ayoub Fentis, EMSI, Morocco
Badri Tijane, EMSI, Morocco
Beverley Cook Essa Hezzam, Nottingham Trent University, UK
Brahim Lejdel, University of El-Oued, Algeria
Brahim Raouyane, Hassan II University – Casablanca Morocco
v
      
Technical  Committee  
El Mehdi Bendriss, Vinci, Morocco
El Moukhtar Zemmouri, Moulay Ismail University, Meknes - Morocco
Emmanuel Simeu, Grenoble-Alpes University, France
Eman Alatawi, Taibah University, Saudi Arabia
Fabrice Monteiro, Lorraine University, France
Faisal Saeed, Taibah University, Kingdom of Saudi Arabia
Fathey Mohammed, Universiti Utara Malaysia (UUM), Malaysia
Fethi Ghazouani, Sorbonne University, France
Firdaous Marzouk, EMSI, Morocco
Funminiyi Olajide, Nottingham Trent University, UK
Georgina Cosma, Nottingham Trent University, UK
Hassna Bensag, EMSI, Morocco
Haithem Mezni, University of Jendouba, Tunisia
Hanin Alahmadi, Taibah University, Saudi Arabia
Hicham Behja, Hassan II University – Casablanca Morocco
Imane Daoudi, Hassan II University – Casablanca Morocco
Jalil Ur Rehman Kazim, University of Glasgow, UK
Jawad Alkhateeb, Taibah University, Saudi Arabia
Kamal Eddine El Khadri, Abdelmalek Essadi University – Tangier Morocco
Khadija Bousmar, EMSI, Morocco
Khalid ELFAHSSI, Sidi Mohamed Benabdellah University – Fez Morocco
Khalid Moussaid, Hassan II University – Casablanca Morocco
Lamiaa Karim, Hassan I University, Settat - Morocco
Lamiae Demraoui, EMSI, Morocco
Leila Fejtah, Hassan II University – Casablanca - Morocco
Maxim Shcherbakov, Volgograd State Technical University, Russia.
Meriem Mandar, Hassan I University , Settat - Morocco
Meryem Khouil, EMSI, Morocco
Moez Krichen, Albaha University, Saudi Arabia
Mohamed Atibi, EMSI, Morocco

vi
  
Technical  Committee  
Mohammed Al Ashwal, Volgograd State Technical University, Russia.
Mohammed Alshargabi, Najran University, KSA
Mohammed Alsarem, Taibah University, Kingdom of Saudi Arabia
Mohamemd El Youssfi, Hassan II University – Casablanca Morocco
Mohammed H. Jabreel, Universitat Rovira i Virgili, Spain.
Mohammed Hadwan, Qasim University, Kingdom of Saudi Arabia.
Mohammed Karim Guennoun, EHTP - Morocco
Mohammed Kolib, Taibah University, KSA
Mohammed Lahbi, Hassan II University, Casablanca Morocco
Mohammed Laraqui, Ibnou Zohr University, Agadir Morocco
Mohammed Rachdi, Hassan II University, Casablanca Morocco
Mohammed Réda Chbihi Louhdi, Hassan II University, Casablanca Morocco
Mohammed Rida, Hassan II university – Casablanca Morocco
Mohammed Talea, Hassan II University – Casablanca Morocco
Mostafa Bellafkih, INPT, Rabat - Morocco
Mostafa Belmakki, INPT, Rabat - Morocco
Mouhamad Chehaitly, Lorraine University, France
Mounia Abik, ENSIAS, Rabat - Morocco
Mouna Selmi, University of Manouba, Tunisia
Nabil Madrane, Hassan II University – Casablanca Morocco
Nadhmi Gazem, Taibah University, Kingdom of Saudi Arabia
Nadia Saber, EMSI, Morocco
Najib el kamoun, Chouaib doukali University - El Jadida Morocco
Noorminshah Iahad, Universiti Teknologi Malaysia, Malaysia
Noura Aknin, Abdelmalek Essadi University – Tangier Morocco
Noureddine Abghour, Hassan II University – Casablanca Morocco
Ouail Ouchetto, Hassan II University - Casablanca, Morocco
Rabab Chakhmoune, Brams, Morocco
Rachid Dehbi, Hassan II University – Casablanca Morocco
Rachid Ouled Lhaj Thami, ENSIAS, Rabat - Morocco
Rachid Saadane, EHTP, Casablanca - Morocco

vii
      
Technical  Committee  
Rashiq Rafiq, Taibah University, KSA
Redouane Kanazy, EMSI, Morocco
Reem Kateb, Taibah University, Saudi Arabia
Rim Koulali, Hassan II University – Casablanca Morocco
Said Jai Andaloussi, Hassan II university – Casablanca Morocco
Said Ouatik El Alaoui, Ibn Tofeil University – Kenitra Morocco
Sanaa Elfilali, Hassan II university, Casablanca Morocco
Sana Ullah Jan, University of West of Scotland, UK
Slim Lammoun, ENSIT, Tunis university
Sofien Dellagi, Lorraine University, France
Soukaine ELhassnaoui, EMSI, Morocco
Stephen Clark, Nottingham Trent University, UK
Taha Hussein, Universiti Malaysia Pahang, Malaysia
Tarik Nahal, Hassan II University – Casablanca Morocco
Tawfik Al-Hadhrami, Nottingham Trent University, UK
Wadii Boulila, University of Manouba, Tunisia
Yousef Fazea, Universiti Utara Malaysia (UUM), Malaysia
Youssef Baddi, Chouaib doukali University - El Jadida Morocco
Zakaria Sabiri, ENSEA, France
Zineb Besri, Abdelmalek Saadi University, Tanger - Morocco
Zouhaier Ben Rabeh, CNCT, Tunisia

viii
  
        Messages  
CONFERENCE HONORARY CHAIR
MESSAGE
I would like to welcome you to the 2nd International
Conference of Advanced Computing and Informatics
(ICACIN 2021) that is held online in Hassan II
University on May 24-25, 2021 and jointly organized
by Faculty of Sciences Ain Chock Casablanca,
Hassan II University and the Yemeni Scientists
Research Group (YSRG) in collaboration with
Information Service Systems and Innovation
Research Group (ISSIRG) in Universiti Teknologi
Malaysia (Malaysia), Data Science Research Group, College of Computer Science and
Engineering, Taibah University (Saudi Arabia) and Higher Normal School of Casablanca.
ICACIN 2021 received 71 submissions from 20 countries including Algeria, China,
France, Iraq, Italy, Jordan, Latvia, Malaysia, Morocco, Nigeria, Pakistan, Philippines,
Rwanda, Saudi Arabia, Senegal, Sri Lanka, Tunisia, United Arab Emirates, United
Kingdom and United States. Of these 71 submissions, 45 submissions have been selected
to be included in the conference proceedings, that will be published in Springer Book
Series: “Advances in Intelligent Systems and Computing”. The ICACIN 2021 program
involves five keynote speeches and five papers presentation sessions during the two days
of the conference.
Thank you to all authors for the participation in ICACIN 2021. Many thanks to
our keynote speakers: Prof. David Brown, Prof Mohamed Younis, Prof. Ts. Dr. Ali
Selamat, Prof Mohamed Younis and Prof. Wadee Alhalabi. Special thanks to the
organizing and technical committees who worked very hard to organize this conference
in this difficult situation.
I wish you every success in your research. Let us pray for the safety of everyone.
Prof. Saddiqi Omar,
Honorary Co-Chair,
Dean of the Faculty of Science Ain Chock,
Hassan II University – Casablanca, Morocco

ix
  
        Messages  
INTERNATIONAL
COLLABORATORS MESSAGE
I am honored to welcome all participants and
guests of the 2nd International Conference of
Advanced Computing and Informatics (ICACIN
2021) that is conduced online at the Faculty of
Science Ain Chock, Hassan II University.

Our team from Taibah University is pleased to

collaborate and work with this great team from Morocco, Yemen and Malaysia.
The theme of ICACIN 2021 conference aims to exploit the advantages of
intelligent computing to serve our communities and countries. The program
includes five keynote speeches that presents interesting research topics on health
informatics using machine learning, internet of things, smart computing and
virtual reality.

I would like to express my sincere appreciation to the organizing and technical

committee members for their great efforts to organize this successful conference.
Many thanks to the keynote speakers and all participants of ICACIN 2021
conference for sharing your research and expertise with us.

I hope you are enjoying the conference and stay healthy.

Dr. Fahad Gabban

Deputy Dean, Research and Postgraduate Studies,
College of Computer Science and Engineering
Taibah University, Saudi Arabia

x
  
        Messages  

CONFERENCE GENERAL CO-CHAIR

MESSAGE
I would like to welcome all of you to the 2nd
International Conference of Advanced Computing
and Informatics (ICACIN 2021).
We are very pleased and honored to continue the
great collaboration for the second time with our
colleagues in Faculty of Sciences Ain Chock
Casablanca at Hassan II University. Our organizing committee includes several
international researchers from Morocco, Yemen, Saudi Arabia, United Kingdom
and Malaysia.
The conference proceedings includes 45 papers that discuss several research
topics such as artificial intelligence, data science, internet of things (IoT),
networking and could computing, information security and computational
informatics. The ICACIN 2021 main theme is: Advances on Smart Computing
and Informatics. All the accepted and presented papers will be published in
Springer book series “Advances in Intelligent Systems and Computing”. The
conference program includes five keynote speeches and five sessions for the two
days.
I would like to express our appreciations to all authors and the keynote speakers
for sharing their expertise and contributions with us. And I would like to thank
the organizing committee for their great efforts in managing the conference; the
technical committee for reviewing all the submitted papers; and Prof. Dr. Janusz
Kacprzyk and Dr. Thomas Ditzinger from Springer.
Finally, we thank all the participants of ICACIN 2021 and hope to see you again
in the next conference.

Assoc. Prof. Dr. Faisal Saeed

ICACIN 2021 General Co-Chair
Taibah University, Saudi Arabia

xi
  
        Messages  

CONFERENCE GENERAL CO-CHAIR

MESSAGE

I am happy to welcome you to the second edition

of the international conference in computer science
and advanced computing organized at the Faculty of
Sciences Ain Chock Casablanca. The ICACIN 2021
conference offers an excellent exchange and meet-
ing opportunity for researchers and industrialists
from different institutions and organizations. This
will undoubtedly allow the sharing of experiences and innovations in the field of
new technologies and communications.
The conference organization is subject to international cooperation between sev-
eral institutions and organizations including the Faculty Of Sciences Ain Chok,
Moroccan association for the promotion of scientific research, Yemeni Scientists
Research Group (YSRG), Service Systems and Innovation Research Group
(ISSIRG) in Universiti Teknologi Malaysia (Malaysia) and Data Science Re-
search Group in Taibah University (Kingdom of Saudi Arabia). The international
contribution of the conference also shows in the diversity of the nationalities of
the participants (20 countries).

I would like to thank the organizing committee for their efforts to make the event
a success in these difficult circumstances. I also thank the keynotes Prof. David
Brown, Prof Mohamed Younis, Prof. Ts. Dr. Ali Selamat, Prof Mohamed Younis and
Prof. Wadee Alhalabi who agreed to share their great experiences with the partici-
pants.

Dr Mohammed Errais,
ICACIN 2021 General Co-Chair,
Faculty of Science Ain Chock,
Hassan II University of Casablanca
Morocco

xii
  
        Messages  

CONFERENCE PROGRAM CO-CHAIR

MESSAGE
It is my pleasure to welcome you all; scientists,
academics, researchers and students from all over
the world to the 2nd International Conference of
Advanced Computing and Informatics (ICACIN
2021). The conference is jointly organized by
several leading universities from different countries
from Morocco, Yemen, Saudi Arabia and Malaysia.
I would like also to take this opportunity to express
my utmost gratitude and sincere thanks and
appreciation to the organizing committee members
for their countless efforts and to the technical committee members for helping in
the review process. Special thanks to the external reviewers for providing
constructive and valuable suggestions for the authors to improve their work.
I also appreciate and thank all authors for submitting and presenting their
papers during the conference especially in these days when COVID 19 still
breakout over the world.
Also, I would like to extend special thanks and appreciation to all who
participated, attended and supported this conference and make it a reality.
Finally, the success of this conference will inspire us more in future .

Assoc. Prof. Dr. Mohammed Al-Sarem,

ICACIN 2021 Program Co-Chair
Taibah University, Saudi Arabia

xiii
  

Keynote Speaker I:
Prof. David Brown
Nottingham Trent University, UK

Keynote title:
“Using machine learning to predict and diffuse
meltdown events in students with autism”

Brief Profile
Professor David Brown is currently Professor in Interactive Systems for Social In-
clusion in School of Science & Technology in Nothingham Trend University. He is
a member of the Interactive Systems Research Group (ISRG). His research interests
include : The design, development and evaluation of Virtual Reality and Multimedia
Learning Environments to promote social inclusion (Adapt, DLTR, NLCB, NLSC,
NCH, NLEA), guidelines for accessible web design for people with a disability
(Mental Health Foundation, Leonardo da Vinci), design guidelines for higher educa-
tion eLearning systems (NTU SDG), design of virtual tutoring agents for people with
cognitive impairments (ESRC), Research in best practice eLearning systems to pro-
mote social inclusion (NTU SDG), User sensitive methodologies for the development
of assistive and adaptive devices for use by people with physical and cognitive im-
pairments (EPSRC, Equal), Pedagogy and virtual environments (ESRC), game-based
learning to promote the engagement of young disaffected students in e-learning (EU
Grundtvig, ESF West Midlands, Derby Libraries, NLSC).
Current and recent research is being conducted with the collaboration, funding and /
or support of the following: The Division of Rehabilitation and Ageing, University
of Nottingham, The Birmingham Institute for the Deaf (BID), Greenhat Interactive:
Social Enterprise that develops and trains in eLearning programmes and materials for
accessible, anytime, anywhere learning for socially excluded groups and disadvan-
taged communities, Information Society Open to All (E-Isotis), Greece, European
Pole of Knowledge (Europole), Italy, Integra: a non-government organisation, part
of the larger Integra-network of partnering agencies, involved in social and economic
development of the countries in Central and Eastern Europe, Bulgaria.

xiv
  

Keynote Speaker II:

Prof Mohamed Younis,

University of Maryland Baltimore County.
United States

Keynote title:
“iRoad: Internet of Radio-equipped On-road and Vehicles-carried Agile
Devices”

Brief Profile
Dr Younis is currently a professor in the department of computer science and electri-
cal engineering at the university of Maryland Baltimore County (UMBC). Before
joining UMBC, he was with the Advanced Systems Technology Group, an Aero-
space Electronic Systems R&D organization of Honeywell International Inc. While
at Honeywell he led multiple projects for building integrated fault tolerant avionics
and dependable computing infrastructure. He also participated in the development of
the Redundancy Management System, which is a key component of the Vehicle and
Mission Computer for NASA’s X-33 space launch vehicle. Dr. Younis’ technical
interest includes network architectures and protocols, wireless networks, embedded
systems, fault tolerant computing, secure communication and distributed real-time
systems. He has published about 300 technical papers in refereed conferences and
journals. Dr. Younis has seven granted and three pending patents. In addition, he
serves/served on the editorial board of multiple journals and the organizing and tech-
nical program committees of numerous conferences. Dr. Younis is a senior member
of the IEEE and the IEEE communications society.

xv
  
Keynote Speaker III:

Prof. Ts. Dr. Ali Selamat,

Universiti Teknologi Malaysia, Malaysia

Keynote title:
“Recent research on detection of vulnerable
plaque in virtual histology intravascular
ultrasound images using machine learning algorithms”

Brief Profile
Prof. Ali Selamat is currently a Dean of Malaysia Japan International Institute of
Technology (MJIIT) which is an education institute that is established by Ministry of
Higher Education Malaysia with the aim of enhancing Japanese Oriented Engineer-
ing Education in Malaysia and Asian with the support from Government of Japan
through Japanese International Cooperation agency (JICA) and Universiti Teknologi
Malaysia together with 29 Japanese University Consortium (JUC). Prior to that, he
was a Chief Information Officer (CIO) and a Director of Center of Communication
and Information Technology, UTM. He is currently elected as a Chair of IEEE Com-
puter Society, Malaysia Section under the Institute of Electrical and Electronics En-
gineers (IEEE), USA and also a member of Malaysia Engineering Deans Council. He
was previously assuming the position of research Dean on Knowledge Economy Re-
search Alliance, UTM. Prof. Ali Selamat has received a [Link]. (Hons.) in IT from
Teesside University, U.K. and [Link]. in Distributed Multimedia Interactive Systems
from Lancaster University, U.K. in 1997 and 1998, respectively. He has received a
Dr. Eng. degree from Osaka Prefecture University, Japan in 2003. He was a principal
consultant of Big Data Analytics, Ministry of Higher Education, Malaysia 2017 and
a keynote speaker in many international conferences. He was a visiting professor at
Kuwait University, and few other universities in Japan, Kingdom of Saudi Arabia,
and Indonesia. Currently he is a visiting professor at Hradec-Kralove University,
Czech Republic and Currently, he is serving as the Editorial Boards of International
Journal of Knowledge Based Systems Elsevier, Netherlands, International Journal of
Information and Database Systems (IJIIDS) under Inderscience Publications, Swit-
zerland and Vietnam Journal of Computer Science under Springer Publications.

xvi
  
Keynote Speaker IV:

Prof. Salil Kanhere

The University of New South Wales, Aus-
tralia

Keynote title:
“Who Are You? New Approaches for Authen-
tication in Smart Spaces”

Brief Profile
Salil Kanhere received the M.S. and Ph.D. degrees from Drexel University, Philadel-
phia, USA. He is currently a Professor of Computer Science and Engineering with
UNSW Sydney, Australia. His research interests include the Internet of Things, cyber
physical systems, blockchain, pervasive computing, cybersecurity, and applied ma-
chine learning. Salil is also affiliated with CISRO’s Data61 and the Cybersecurity
Cooperative Research Centre. He is a Senior Member of the IEEE and ACM and an
ACM Distinguished Speaker. He has received the Friedrich Wilhelm Bessel Re-
search Award (2020) and the Humboldt Research Fellowship (2014), both from the
Alexander von Humboldt Foundation in Germany. He has held visiting positions at
I2R Singapore, Technical University Darmstadt, University of Zurich and Graz Uni-
versity of Technology. He serves as the Editor in Chief of the Ad Hoc Networks
journal and as an Associate Editor of the IEEE Transactions On Network and Service
Management, Computer Communications, and Pervasive and Mobile Computing. He
has served on the organising committee of several IEEE/ACM international confer-
ences and is the General Chair for the IEEE International Conference on Blockchain
and Cryptocurrency (ICBC) 2021. He has co-authored a book titled Blockchain for
Cyberphysical Systems published by Artech House in 2020.

xvii
  
Keynote Speaker V:

Prof. Wadee Alhalabi,

King Abdulaziz University, Saudi Arabia

Keynote title:
“Virtual Reality in Health Applications”

Brief Profile
Professor of Artificial Intelligence, Computer Science Department, Faculty of Com-
puting and Information Technology Member of the university promotion committee,
Former Member of the university scientific council, Lead of the Virtual Reality Re-
search Group and Director of the House of Expertise of Virtual Reality and Artificial
Intelligence.
Education:
2008 University of Miami, Fl, USA, PhD in Electrical and computer engineering
2004 University of Miami, Fl, USA MSECE in Electrical and computer engineering
1996 Umm Alqura university BEng. in Electrical and Computer Engineering
Research Interest: Artificial Intelligence and Machine Learning, Image processing
and Virtual reality

xviii
  

Conference Program & Presentation Schedule

Day 1: Monday, 24th May, 2021

Zoom Meeting ID: 832 2599 8622

Zoom Passcode: ICACIn21
8:00 am - 8:10 am Doa’a Recitation
(GMT time)
8:10 am - 8:20 am Welcoming speech by Dr. Mohammed Errais, the
ICACIN 2021 General Co-Chair
8:20 am - 8:30 am Welcoming speech by Dr. Faisal Saeed, the ICACIN
2021 General Co-Chair
8:30 am – 8:50 am Speech and officially opening ICACIN 2021 by Prof.
Saddiqi Omar, Honorary ICACIN 2021 Chair
8:50 am–9:30 am Keynote Speech I:
Prof. David Brown. Nothingham Trend University,
United Kingdom

Title: Using machine learning to predict and diffuse

meltdown events in students with autism
9:30 am – 10:10 am Keynote Speech II:
Prof Mohamed Younis, University of Maryland Balti-
more County. United States

Title: iRoad: Internet of Radio-equipped On-road and

Vehicles-carried Agile Devices

xix
  

10:10 am– 10:50 am Keynote Speech III:

Prof. Ts. Dr. Ali Selamat, Universiti Teknologi Ma-
laysia, Malaysia

Title: Recent research on detection of vulnerable

plaque in virtual histology intravascular ultrasound
images using machine learning algorithms  
10:50 am-11:00 am Break

Session I: 11:00 am – 1:15 pm (GMT time)

Zoom Meeting ID: 832 2599 8622

Zoom Passcode : ICACIn21

Artificial Intelligence I

Adaptive Learning Content based on Learning

Styles in Learning Management System
11:00 pm – 11:15 am E.H Lim, Koo Yuen Phan, Tien Fui Yong, Yin Ping
Yeck, Manoranjitham Muniandy and Anbuselvan
Sangodiah

A model-based approach for COVID-19 propaga-

tion
11:15 am - 11:30 am
Abdeltif Elbyed, Isaac Doku, Tarik Nahhal and Adoni
Wilfried Yves Hamilton

Numerical modeling of the mechanical behavior of

recycled flexible PVC loaded with different per-
centages of chicken feathers
11:30 am - 11:45 am
Abdelghani Lakhdar, Aziz Moumen, Zineb Laabid
and Khalifa Mansouri

xx
  

Covid-19 detection from Pneumonia image classifi-

cation using Deep Learning
11:45 am – 12:00 pm
Abdeltif Elbyed and Mikhaël Presley Kibinda-
Moukengue

COVID-19 Infected Regions segmentation in Lung

CT Images based on Deep Learning and Bidimen-
12:00 pm – 12:15 pm sional Emperical Mode Decomposition
Said Jai Andaloussi and Ouail Ouchetto

Towards the Prediction of the Elasticity of Bio

Loaded Polypropylene Using the Artificial Neural
Networks
12:15 pm – 12:30 pm
Zineb Laabid, Aziz Moumen, Abdelghani Lakhdar
and Khalifa Mansouri

Post covid-19: Deep image processing AI to analyze

social distancing in a human community
12:30 pm – 12:45 pm Francis Class-Peters, Wilfried Yves Hamilton
Adoni, Nahhal Tarik, Abdeltif Elbyed, Moez Krichen,
Charles Kimpolo and Franck Kalala Mutombo

Design and implementation of ImALeG serious

game: Behavior of Non-Playable Characters (NPC)
12:45 pm – 1:00 pm
Yassine Tazouti, Siham Boulaknadel and Youssef
Fakhri

DETECTION OF PERSONALITY TRAITS

THROUGH HANDWRITING ANALYSIS USING
MACHINE LEARNING APPROACH
1:00 pm – 1:15 pm
Sajeevan Srikumar and Wiraj Udara Wickramaarach-
chi

xxi
  

Session II: 2:00 pm – 5:30 pm (GMT time)

Zoom Meeting ID: 832 2599 8622
Zoom Passcode: ICACIn21

Data Science

Data Analytics Model for Home Improvement

Store
2:00 pm – 2:15 pm
Jia Yeng Lim, Mohd Norshahriel Abd
Rani and Nabilah Filzah Mohd Radzuan

Intelligent Models for Mining Social Media Data

Sultan Albloushi, Mohamedammar Alnuaimi, Omran
2:15 pm - 2:30 pm
Alhosani, Hassan Al Ali, Khalid Almarashda and Sa-
naa Elyassami

A comparative study of Alzheimer disease stage

classification with Transfer Learning
2:30 pm - 2:45 pm
Tariq Saeed

Designing an Arabic Google Play Store User Re-

view Dataset for Detecting App Requirement Issues
2:45 pm - 3:00 pm
Maha Alshamani, Mohammed Al-Sarem, Faisal Saeed
and Wafa'A Almutairi

A Survey of Machine Learning and Deep Learning

3:00 pm - 3:15 pm Applications in Genome Editing
Talal Almutiri, Faisal Saeed and Manar Alassaf

xxii
  

Educational Data Mining techniques for detecting

undesirable students’ behaviors and predicting stu-
3:15 pm - 3:30 pm dents’ performance: A comparative study
Imane El Mourabit, Said Jai Andaloussi and Nored-
dine Abghour

COVID-19: Analysis and measurement of the influ-

ence of the tweets to helping the public health sec-
3:30 pm - 3:45 pm tor to fight Coronavirus
Abdessamad Essaidi, Dounia Zaidouni and Mostafa
Bellafkih

Predicting the e-signing likelihood of loan using

machine learning models combining clustering with
3:45 pm – 4:00 pm
supervised learning
Abdellatif Aattouf, Said El Kafhali and Youssef Saadi

Hybrid Students’ Academic Performance and

Dropout Prediction Models Using Recursive Fea-
4:00 pm – 4:15 pm ture Elimination Technique
Moshood Alarape, Ahmed Ameen and Kayode Ade-
wole

Cloud computing & Networking

Adoption of Docker Containers as an Infrastruc-

ture for Deploying Software Applications: a review

4:15 pm - 4:30 pm W.M.C.J.T. Kithulwatta, Wiraj Udara Wickramaar-

achchi, Kpn Jayasena, Btgs Kumara and Rmkt Rathna-
yaka

xxiii
  

A Parallel Genetic Algorithm for solving the Vehi-

cle Routing Problem with Drone medication deliv-
4:30 pm - 4:45 pm ery
Ouiss Mustapha, Ettaoufik Abdelaziz, Marzak Ab-
delaziz and Tragha Abderrahim

Review on Blockchain & Access Control Systems

4:45 pm - 5:00 pm
Mohamed Anass Amallah and Noreddine Abghour

Routing and charging of electric vehicles: Over-

5:00 pm - 5:15 pm view
Meriem Abid and Mohamed Tabaa

Study of the impact of the profoundness of tunnel

in the context of transition from IPv4 to IPv6 at the
5:15 pm – 5:30 pm
local, regional, and national levels.
Dounia El Idrissi, Najib Elkamuon and Rachid Hilal

Day 2: Tuesday, 25th May, 2021

Zoom Meeting ID 849 6754 6258
Zoom Passcode : ICACIn21
8:00 am - 8:40 am Keynote Speech IV
(GMT time) Prof. Salil Kanhere , The University of New South
Wales, Australia

Title: Who Are You? New Approaches for Authenti-

cation in Smart Spaces

xxiv
  

8:40 am - 9:20 am Keynote Speech V

Prof. Wadee Alhalabi, King Abdulaziz University,
Saudi Arabia

Title: Virtual Reality in Health Applications

9:20 am – 9:30 am Break

Session I: 9:30 am – 11:00 am (GMT time)

Zoom Meeting ID 849 6754 6258
Zoom Passcode : ICACIn21

Artificial Intelligence II

9:30 am - 9:45 am A Face Mask Detection System Based on Deep

Learning Convolutional Neural Networks
Pierre Stanislas Birame Ndong, Wilfried Yves Hamil-
ton Adoni, Nahhal Tarik, Abdeltif Elbyed, Moez
Krichen, Charles Kimpolo and Franck Kalala
Mutombo

Implementing Weka deep learning algorithm for

detecting COVID-19 cases using extracted features
from XCR and chest CT images
9:45 am – 10:00 am
Anmar Abuhamdah, Ghaith Jaradat and Mutasem Al-
Smadi

Performance comparison of K-Nearest Neighbor,

Random Forest, and Multiple Linear Regression to
10:00 am - 10:15 am predict photovoltaic panels' power output
Souhaila Chahboun and Mohamed Maaroufi

xxv
  

Voting Classifier-based Intrusion Detection for IoT

Networks
10:15 am - 10:30 am Muhammad Almas Khan, Muazzam A Khan, Shahid
Latif, Awais Aziz Shah, Mujeeb Ur Rehman, Wadii
Boulila, Maha Driss and Jawad Ahmad

Glazing's Potential Energy Savings for Green

Buildings
10:30 am - 10:45 am
Badr Chegari, Mohamed Tabaa, Emmanuel Simeu,
Fouad Moutaouakkil and Hicham Medromi

Regeneration a Graphical User Interface Using

Deep Learning
10:45 am – 11:00 am
Cheddi Fatima, Ahmed Tahiri and Mohammed Ser-
rhini

Session II: 11:00 am - 12:30 pm (GMT time)

Zoom Meeting ID: 849 6754 6258
Zoom Passcode: ICACIn21

Information Security
A Review of Blockchain and the Benefits for Digital
Marketing-related Applications of Blockchain Inte-
11:00 am – 11:15 am gration
Abirou Mouhssine and Abghour Noureddine

Secure Remote Workplace 4EM model

11:15 am – 11:30 am
Sarma Cakula

Review in Authentication for Mobile Ad hoc Net-

11:30 am – 11:45 am work

xxvi
  

Hafida Khalfaoui

A Panoramic Evaluation of Machine Learning and

Deep Learning Aided Ransomware Detection Tools
11:45 am – 12:00 pm Using a Hybrid Cluster of Rich Smartphone Traits
Hiba Zuhair

A Big Data driven model for secured systems: A

Blockchain-based architecture for Cloud Compu-
ting and IoT security
12:00 pm – 12:15 pm
Ndayikengurukiye Aristide, Omary Fouzia and Abder-
rahmane Ez-Zahout

Hybrid Filter and Wrapper Feature Selection

Method for Enhancing Detection Process of Phish-
ing Websites
12:15 pm – 12:30 pm Wafa'A Almutairi, Faisal Saeed, Mohammed Al-
Sarem and Maha Alshamani

Session III: 1:00 pm - 3:30 pm (GMT time)

Zoom Meeting ID 849 6754 6258
Zoom Passcode : ICACIn21

Internet of Things

E-Business as Business Continuity plan in the Du-

ration of Covid-19: The Business Case of Saudi
1:00 pm – 1:15 pm Arabia
Eman Alahsnqiti, Alzahraa Alsharif, Maha Al-
shamani and Abdelrahman Karrar

xxvii
  

Heuristic methods for the container stacking prob-

1:15 pm – 1:30 pm lems : A review
Farah Al Kaderi, Koulali Rim and Mohamed Rida

A Smart Power Grid architecture for microneigh-

borhood based on Green IoT
1:30 pm – 1:45 pm
Asmae Chakir, Mohamed Tabaa, Fouad Moutaouak-
kil, Hicham Medromi and Karim Alami

Study of LoRaWAN performance under a noisy in-

dustrial channel
1:45 pm – 2:00 pm
Nisrine Bahri, Mohamed Tabaa, Safa Saadaoui, Mo-
hamed Sadik and Hicham Medromi

Weight Initialization Techniques for Deep Learn-

ing Algorithms in Remote Sensing: Recent Trends
2:00 pm – 2:15 pm and Future Perspectives
Wadii Boulila, Maha Driss, Eman Alshanqiti, Mo-
hamed Al-Sarem, Faisal Saeed and Moez Krichen

Optimization of Wind Farms Based on the Selec-

tion of Types of Wind Turbine
2:15 pm – 2:30 pm
Bellat Abdelouahad, Mansouri Khalifa and Raihani
Abdelhadi

Optimization of Noisy Gradient Descent Bit-Flip

Decoding for LDPC Codes parameters based on
2:30 pm– 2:45 pm Taguchi method
Hajar Elouakili, Rachid El Alami, Mehdaoui
Youness, Driss Chenouni and Zakia Lakhliai

xxviii
  

Deep neural networks hyperparameter optimiza-

tion using particle swarm optimization for detect-
2:45 pm – 3:00 pm ing frauds transactions
Mohammed Tayebi and Said El Kafhali

EmPRM: A novel Multi-agent model for modeling

and simulating of electromagnetic waves using
3:00 pm – 3:15 pm Netlogo Platform
Hamid Bezzout and Hanan El Faylali

Fighting Covid-19: what role for digital technol-

3:15 pm – 3:30 pm ogy?
Ouail Ouchetto and Said Jai-Andalouss

3:30 pm – 3:45 pm Break

3:45 pm – 4:00 pm Closing Ceremony

xxix
    
Table  of  Contents  

Organizing Committee ii
Technical Committee ix
Messages ix
Keynote Speakers xiv

Conference Program and Sessions Schedule xix

Artificial Intelligence I 1
Adaptive Learning Content based on Learning Styles in Learn- 2
ing Management System
Lim Ean Heng, Phan Koo Yuen, Yong Tien Fui, Manoranjitham
Muniandy, Anbuselvan Sangodiah and Yeck Yin Ping
A model-based approach for COVID-19 propagation 3
Abdeltif El Byed, Isaac Doku, Tarik Nahhal, Adoni Wilfried Yves
Hamilton
Numerical modeling of the mechanical behavior of recycled 4
flexible PVC loaded with different percentages of chicken
feathers
Abdelghani Lakhdar, Aziz Moumen, Zineb Laabid and Khalifa
Mansouri
Covid-19 detection from Pneumonia image classification using 5
Deep Learning
Abdeltif EL BYED1 and Mikhaël Presley KIBINDA-
MOUKENGUE
COVID-19 Infected Regions segmentation in Lung CT Images 6
based on Deep Learning and Bidimensional Emperical Mode
Decomposition
Said Jai Andaloussi and Ouail Ouchetto

xxx
    
Table  of  Contents  
Towards the Prediction of the Elasticity of Bio Loaded Poly- 7
propylene Using Artificial Neural Networks
Zineb Laabid, Aziz Moumen, Abdelghani Lakhdar and Khalifa
Mansouri
Post covid-19: Deep image processing AI to analyze social dis- 8
tancing in a human community
Francis Class-Peters, Wilfried Yves Hamilton Adoni, Tarik
Nahhal, Abdeltif El Byed, Moez Krichen , Charles Kimpolo, and
Franck Mutombo
Design and implementation of ImALeG serious game: Behav- 9
ior of Non-Playable Characters (NPC)
Yassine Tazouti; SihamBoulaknadel; Youssef Fakhri
Detection of Personality Traits through Handwriting Analysis 10
Using Machine Learning Approach
S. Sajeevan and Wiraj Udara Wickramaarachchi
11
Data Science
Hybrid Students’ Academic Performance and Dropout Predic- 12
tion Models Using Recursive Feature Elimination Technique
Moshood A. Alarape, Ahmed O. Ameen and Kayode S. Adewole
Stage classification of Alzheimer’s disease using Transfer 13
Learning
Tariq Saeed Mian
Predicting the e-signing likelihood of loan using machine learn- 14
ing models combining clustering with supervised learning
Abdellatif Aattouf , Said El Kafhali and Youssef Saadi
Designing an Arabic Google Play Store User Review Dataset 15
for Detecting App Requirement Issues
xxxi
    
Table  of  Contents  
Maha Al-Shamani, Mohammed Al-Sarem, Faisal Saeed, and
Wafa'a
A Survey of Machine Learning and Deep Learning applications 16
in Genome Editing
Talal Almutiri, Faisal Saeed and Manar Alassaf
Educational Data Mining techniques for detecting undesirable 17
students’ behaviors and predicting students’ performance: A
comparative study
Imane EL MOURABIT and Said JAI ANDALOUSSI and Nored-
dine ABGHOUR
COVID-19: Analysis and measurement of the influence of the 18
tweets to help the public health sector to fight Coronavirus
Abdessamad Essaidi, Dounia Zaidouni, and Mostafa Bellafkih
Data Analytics Model for Home Improvement Store 19
Lim Jia Yeng, Mohd Norshahriel Abd Rani, and Nabilah Filzah
Mohd Radzuan
Intelligent Models for Mining Social Media Data 20
Sanaa Elyassami, Sultan Albloushi, MohamedAmmar Alnuaimi,
Omran Alhosani, Hassan Al Ali, Khalid Almarashda
Cloud computing & Networking 21
Routing and charging of electric vehicles: survey 22
Meryem ABID1 and Mohamed TABAA
A Parallel Genetic Algorithm for solving the Vehicle Routing 23
Problem with Drone medication delivery
Mustapha OUISS, Abdelaziz ETTAOUFIK, Abdelaziz MARZAK,
Abderrahim TRAGHA
Review on Blockchain & Access Control Systems 24

xxxii
    
Table  of  Contents  
AMALLAH Mohamed Anass, ABGHOUR Noreddine,
MOUSSAID Khalid, EL OMRI Amina, RIDA Mohamed
Adoption of Docker Containers as an Infrastructure for De- 25
ploying Software Applications: a review
W.M.C.J.T. Kithulwatta, Wiraj Udara Wickramaarachchi, K.P.N.
Jayasena, B.T.G.S. Kumara and R.M.K.T. Rathnayaka
Study of the impact of the profoundness of tunnel in the con- 26
text of transition from IPv4 to IPv6 at the local, regional, and
national levels
Dounia EL IDRISSI, Najib ELKAMOUN ,and Rachid HILAL
Artificial Intelligence II 27
A Face Mask Detection System Based on Deep Learning Con- 28
volutional Neural Networks
Pierre Stanislas Birame Ndong, Wilfried Yves Hamilton Adoni,
Tarik Nahhal, Charles Mberi Kimpolo, Moez Krichen, Elbyed Ab-
deltif, Ismail Assayad, and Franck Kalala
Deep learning algorithm for detecting COVID-19 cases using 29
extracted features from XCR and chest CT images
Anmar Abuhamdah, Ghaith Jaradat and Mutasem Al-Smadi
Performance comparison of K-Nearest Neighbor, Random For- 30
est, and Multiple Linear Regression to predict photovoltaic
panels' power output
CHAHBOUN Souhaila and MAAROUFI Mohamed
Voting Classifier-based Intrusion Detection for IoT 31
Networks
Muhammad Almas Khan, Muazzam A Khan, Shahid Latif, Awais
Aziz Shah, Mujeeb Ur Rehman, Wadii Boulila, Maha Driss, and
Jawad Ahmad

xxxiii
    
Table  of  Contents  
Glazing's Potential Energy Savings for Green Buildings 32
Badr Chegari, Mohamed Tabaa, Emmanuel Simeu, Fouad
Moutaouakkil, Hicham Medromi
Regenerating a Graphical User Interface Using Deep Learning 33
Fatima CHEDD, Ahmed TAHIRI and Mohammed SERRHINI
Information Security 34
A Review of Blockchain and the Benefits for Digital Marketing- 35
related Applications of Blockchain Integration
Mouhssine Abirou, Noureddine Abghour
Secure Remote Workplace 4EM model 36
Cakula Sarma
Review in Authentication for Mobile Ad hoc Network 37
Khalfaoui Hafida, Farchane Abderrazak, and Safi Said
A Panoramic Evaluation of Machine Learning and Deep 38
Learning Aided Ransomware Detection Tools Using a Hybrid
Cluster of Rich Smartphone Traits
Hiba Zuhair
A Big Data driven model for secured systems: 39
A Blockchain-based architecture for Cloud Computing and
IoT security
Aristide NDAYIKENGURUKIYE, Abderrahmane EZ-ZAHOUT
and Fouzia OMARY
Hybrid Filter and Wrapper Feature Selection Method for En- 40
hancing Detection Process of Phishing Websites
Wafa'a Almutairi, Faisal Saeed, Mohammed Al-Sarem, and Maha
Alshamani

xxxiv
    
Table  of  Contents  
Internet of Things 41
Optimization of Wind Farms Based on the Selection of Types 42
of Wind Turbine
Bellat Abdelouahad, Mansouri Khalifa and Raihani Abdelhadi
Heuristic methods for the container stacking problems : A re- 43
view
Farah Al Kaderi, Koulali Rim and Mohamed Rida

A Smart Power Grid architecture for microneighborhood based 44

on Green IoT
Asmae Chakir, Mohamed Tabaa, Fouad Moutaouakkil, Hicham
Medromi and Karim Alami
Study of LoRaWAN performance under a noisy industrial 45
channel
Nisrine Bahri, Mohamed Tabaa, Safa Saadaoui, Mohamed Sadik
and Hicham Medromi
Weight Initialization Techniques for Deep Learning Algorithms 46
in Remote Sensing: Recent Trends and Future Perspectives
Wadii Boulila, Maha Driss, Eman Alshanqiti, Mohamed Al-Sarem,
Faisal Saeed and Moez Krichen
Toward E-Business during Covid-19: The Business Case of 47
Saudi Arabia
Eman Alahsnqiti, Alzahraa Alsharif, Maha Alshamani and Ab-
delrahman Karrar
Optimization of Noisy Gradient Descent Bit-Flip Decoding for 48
LDPC Codes parameters based on Taguchi method
Hajar Elouakili, Rachid El Alami, Mehdaoui Youness, Driss Cheno-
uni and Zakia Lakhliai

xxxv
    
Table  of  Contents  
Deep neural networks hyperparameter optimization using par- 49
ticle swarm optimization for detecting frauds transactions
Mohammed Tayebi and Said El Kafhali
EmPRM: A novel Multi-agent model for modeling and simulat- 50
ing of electromagnetic waves using Netlogo Platform
Hamid Bezzout and Hanan El Faylali
Fighting Covid-19: what role for digital technology? 51
Ouail Ouchetto and Said Jai-Andalouss

xxxvi
  
Abstracts  

Artificial Intelligence I

1
  
Abstracts  

Adaptive Learning Content based on Learning

Styles in Learning Management System

Lim Ean Heng, Phan Koo Yuen, Yong Tien Fui, Manoranjitham Muniandy,
Anbuselvan Sangodiah and Yeck Yin Ping

Universiti Tunku Abdul Rahman, Kampar 31900, Malaysia

ehlim@[Link], phanky@[Link],
yongtf@[Link], manoranm@[Link],
anbuselvan@[Link], yeckyp@[Link]

Abstract. An adaptive learning content which created specifically

based on student learning style has drawn many research interests in
the field of e-learning. Initially, the adaptive learning content is
started in order to enhance the teaching and learning strategy based
on student needs. The main objective of this paper is to map the proper
learning content with the student learning style based on Felder Sil-
verman learning style model. When the student login into the learning
management system, their learning style will be identified and the
most appropriate learning content which fits their learning style will
be automatically assigned to them. The motivation to this research is
to enhance the learning performance of the student. A total of 150
students have voluntary participated in the experiments conducted in
a lab environment. The results of the experiments indicate that when
a proper learning material is presented to a student based on their
learning styles, it improved their learning performance. The result re-
veals the system effectiveness for which it ap-pears that the proposed
approach may be promising.

Keywords: personalization, adaptive, learning management system,

learning styles.

2
  
Abstracts  

A model-based approach for COVID-19

propagation

Abdeltif El Byed1, Isaac Doku2, Tarik Nahhal1, Adoni Wilfried Yves

Hamilton1
1
Science Faculty of Casablanca, Hassan II University, Morocco
[Link]@[Link]
2
African Institute for Mathematical Sciences (AIMS), Senegal

Abstract. The Corona-virus disease (COVID-19) outbreak in this ar-

ticle is modeled by the Susceptible, Infectious, and Recovered (SIR)
model. The SIR model is estimated by using real-time historical
COVID-19 data provided by the Center for Systems, Science and En-
gineering at Johns Hopkins University (CSSE) on GitHub. Among
the mathematical models for disease spread, the SIR model is a favor-
ite model used by most researchers to predict the transmission of a
pandemic. A time-dependent SIR model is therefore used in this paper
to predict the evolution of COVID 19 in time and forecast the propa-
gation of COVID-19 specifically in Morocco using COVID-19 his-
torical dataset and the autoregressive model with exogenous input
(ARX) or Ridge Regression. According to the SIR model, the infected
cases predicted in Morocco between 27th April and 2nd May were
somewhere between 4000 and 4100 infected people with an estimated
basic reproduction number of 5.83. SIR model and Ridge Regression
is then trained to predict and forecast the infection rate and the recov-
ery rate in Morocco.

Keywords: SIR, Ridge Regression, COVID-19, Transmission rate.

3
  
Abstracts  

Numerical modeling of the mechanical behavior of

recycled flexible PVC loaded with different
percentages of chicken feathers

Abdelghani Lakhdar*, Aziz Moumen, Zineb Laabid and Khalifa Mansouri

SSDIA Laboratory, Hassan II University of Casablanca, ENSET of Mohammedia,

28830, Morocco
lakhdarabdelghani11@[Link],Moumenaziz1@[Link],Laabid
.zineb@[Link] khmansouri@[Link].

Abstract. The complex problems in mechanical subjects for experi-

mental studies still exist, this is why it is necessary to resort to numer-
ical modeling which develops exponentially and follows the evolu-
tion of the computer tool. Modeling carried out by simulating a phys-
ical phenomenon by mathematical equations which will be integrated
into a computer tool to give good results to an analytically persistent
problem. This modeling is presented in this article to study the me-
chanical behaviours of a composite material, we are talking about the
addition of a bio-filler in the form of chicken feathers according to
different percentages 5, 10 and 15% on flexible Polyvinyl chloride
(PVC) already aged. The numerical modeling of this article intervenes
in the validation of the experimental result, since a preliminary exper-
imental study has already been carried out. We use a numerical pro-
cessing model well known for its relevance and the accuracy of the
results, which is the finite element calculation model.
This model is used on three samples of bio-loaded PVC with three
different percentages of chicken feathers. The results obtained are in-
tegrated into a comparative study to find the appropriate percentage
that improves the mechanical characteristics, with the aim of increas-
ing the number of recyclability of this material.

Keywords: Numerical Modeling, Flexible PVC, Chicken Feathers.

4
  
Abstracts  
Covid-19 detection from Pneumonia image
classification using Deep Learning

Abdeltif EL BYED1 and Mikhaël Presley KIBINDA-MOUKENGUE2

1
Science Faculty of Casablanca, Hassan II University, Morocco.
[Link]@[Link]
2
African Institute for Mathematical Sciences, Senegal.
[Link]@[Link]

Abstract. In the face of this COVID-19 pandemic that is raging

around the world, many scientists are trying to find effective solutions
that will stop the spread of this pandemic. In this article, we are inter-
ested in chest X-ray imaging. However, several works have shown
that patients with COVID-19 have some abnormalities in their chest
X-ray images. Detecting these abnormalities represents a great chal-
lenge in clinicians' decision-making. The aim of this paper is the au-
tomated detection of images with abnormalities from chest X-ray im-
ages using deep learning. In this work, two pre-trained models
(VGG16 and DenseNet121) were used to diagnose COVID-19 related
infections. This study shows considerable efficiency in terms of per-
formance results (accuracy, cost, sensitivity, the algorithm used, etc.)
in COVID-19 detection.

Keywords: X-ray imaging, decision-making, pre-trained models,

deep learning.

5
  
Abstracts  
COVID-19 Infected Regions segmentation in Lung
CT Images based on Deep Learning and
Bidimensional Emperical Mode Decomposition

Said JAI-ANDALOUSSI1 and Ouail OUCHETTO2

1
Science Faculty of Casablanca, Hassan II University, Morocco.
2
Faculty of Juridical Sciences, Economic and Social, Casablanca, Hassan II
University, Morocco.
[Link]@[Link],
[Link]@[Link],

Abstract. For COVID-19 patients, the common stage of diagnosis is

computed tomography (CT). Radiologists are often asked to estimate
the degree of damage relative to lung volume. This is a time-consum-
ing process because the radiologist should examine all the axial slices
on the CT and segment them. In this work, we propose an automatic
and effective deep learning system to segment lung CT images of the
COVID-19 infections. Principally, we use a feature attention module
(FAM) that can enhance the quality of segmentation and improve the
model learning capability. The proposed model includes convolution
layers that enlarge the receptive field by using frequency-domain fea-
tures obtained by Bidimensional Empirical Mode Decomposition
(BEMD). This approach is used to decompose an image into a series
of functions called”Bidimensional Mode Intrinsic Function” (BIMF)
and a residual function. The proposed method could allow deep learn-
ing models to learn relevant information and COVID-19 infection-
related characteristics that deliver accurate segmentation results. This
method has been tested on the COVID-19 CT Lung images and In-
fection Segmentation dataset and the results obtained prove the effec-
tiveness of our proposed method.

6
  
Abstracts  
Towards the Prediction of the Elasticity of Bio
Loaded Polypropylene Using Artificial Neural
Networks

Zineb Laabid*, Aziz Moumen, Abdelghani Lakhdar and Khalifa Mansouri

SSDIA Laboratory, Hassan II University of Casablanca, ENSET of
Mohammedia, 28830, Morocco.
[Link]@[Link],Moumenaziz1@[Link],
lakhdarabdelghani11@[Link],
khmansouri@[Link].

Abstract. The determination of the properties of bio composites is

the key subject of current research. The mechanical characteristics of
these new materials depend on several factors, we can note the char-
acteristics of the constituents, the process followed during the manu-
facture etc. Several solutions have been made available to determine
these properties such as experience tests, the methods of homogeni-
zation, and the finite element method. Currently, thanks to artificial
intelligence, smart solutions have been able to give better results, par-
ticularly in the prediction of the mechanical properties of materials.
In this article, we will use artificial neural networks to predict the
Young's modulus of polypropylene loaded at 15% with horn fibers.
Using Mori Tanaka's model, which is a form of the homogenization
method, we will generate a dataset to feed our feed forward back prop-
agation and demonstrate that this bio composite is gaining in terms of
elasticity with a very good value of coefficient regression, 099, and
with a best validation of performance, 0.034737 obtained at epoch 11.

Keywords: Feed forward back propagation, Numerical analysis, Bio

Composites, Young Modulus.

7
  
Abstracts  
Post covid-19: Deep image processing AI to
analyze social distancing in a human community

Francis Class-Peters1, Wilfried Yves Hamilton Adoni2;3 Tarik Nahhal2,

Abdeltif El Byed 2, Moez Krichen 3, Charles Kimpolo1, and Franck
Mutombo Kalala

1 African Institute for Mathematical Sciences, Senegal

2 Science Faculty of Casablanca, Hassan II University, Morocco
3 International university of Casablanca
REDCAD Research Uni

Abstract. In the case of the ght against COVID-19, social distancing

has proven to be an e ective intervention to minimize disease trans-
mis-sion. AI / Deep Learning has recently lent itself as a great tool to
solving almost every conceivable problem in daily life and has shown
signi cant promising results for such problems. In this article, we will
explore in-depth how Python, Computer Vision, and Deep Learning
can be used to track social distances in public and workplaces.
Our study seeks to contribute to the e ort in the ght against the pan-
demic by providing a tool that basically functions as a social distanc-
ing monitoring device. Since social distancing is one of the rst-line,
non-pharmaceutical interventions adopted to combat the COVID-19
pan-demic and the with growing evidence supporting its e ectiveness,
we believe that strict accordance with this policy should be encour-
aged. With the sole purpose of ensuring e ective social distancing in
public places, our tool will analyze real-time and/ or pre-recorded
video feed, detect and compute the pairwise metric distances between
individuals to validate whether or not social distancing is maintained.

Keywords: COVID-19, Deep Learning, CNN, Faster R-CNN, Com-

puter Vision, Social Distancing

8
  
Abstracts  
Design and implementation of ImALeG serious
game: Behavior of Non-Playable Characters
(NPC)

Yassine Tazouti1[0000-0002-6546-6237], SihamBoulaknadel2[0000-0001-7538-7295],

Youssef Fakhri 1[0000-0002-5647-303X]
1
Faculty of science, Ibn Tofail University, Kénitra, Morocco
2
IRCAM, Rabat, Morocco
{[Link], fakhri}@[Link]
boulaknadel@[Link]

Abstract. Non-Playable Character behavior is necessary to make a

faithful simulation situation. In this paper, a computational intelli-
gence method is used to design a behavior models of Non-Playable
Character inside a virtual reality serious game which aims to help peo-
ple learning Amazigh language. The Non-Playable Characters are
modeled using the concept of FuSM. Through a survey, we evaluated
the emotional impact to identify the player’s feelings towards the vir-
tual environment.

Keywords: Behavior design, FuSM, Heuristic methods, Virtual real-

ity, Video games, Amazigh language

9
  
Abstracts  
Detection of Personality Traits through
Handwriting Analysis Using Machine Learning
Approach
S. Sajeevan1 and Wiraj Udara Wickramaarachchi 1, 2
1
Department of Computing, Faculty of Applied Science,
Rajarata University of Sri Lanka, Mihintale, Sri Lanka.
2
School of Computer Science and Technology,
Wuhan University of Technology, Wuhan, Hubei, China.
srisajeeve7@[Link]
wirajudara@[Link]

Abstract. The human emotions and behaviors differ from person to per-
son. Human handwriting is a unique one like finger print, which carries
the richest information to gain the insights into the physical, mental and
emotional state of the writer. The science of analyzing handwriting is
called graphology. Graphological rules are defined by considering the fea-
tures of the handwriting. The proposed approach is carried out with the
intention of finding how successful is the computer aided handwriting
analysis and the most appropriate machine learning classification ap-
proach for graphological feature classification. An application is devel-
oped to intake handwriting samples and to output personality trait of the
writer. The application analyzes 3 major features: pen pressure, page mar-
gin and word size of a handwriting sample. Three hundred handwriting
samples were tested and the outputs were stored as a dataset. The dataset
is used to train and test Support Vector Machine and K-Nearest Neighbor
models. The classification reports depict high accuracies of 96% and 85%
for SVM and KNN models respectively. Application was tested with a
small team of candidates and received good responses. More than 65% of
them were satisfied with the functionality and the output of the applica-
tion. The topic is open for further analyzes one like handwriting sample
can be segmented further into words and characters and can analyze the
features of them.

Keywords: Graphology, Personality Traits, Machine Learning,

SVM, KNN

10
  
Abstracts  

Data Science

11
  
Abstracts  
Hybrid Students’ Academic Performance and
Dropout Prediction Models Using Recursive
Feature Elimination Technique

Moshood A. Alarape1, Ahmed O. Ameen2 and Kayode S. Adewole2

1
Department of Computer Science Federal Polytechnic, Bida, Nigeria
2
Department of Computer Science, University of Ilorin, Ilorin, Nigeria
1
[Link]@[Link],
2
aminamed@[Link],[Link]@unilorin .[Link]

Abstract. Predicting Students’ Academic Performance (SAP) and dropout

are crucial aspects of any academic institution management. It enhances the
close monitoring of students’ academic progress and helps in averting possible
negative consequences of poor students’ performances. Several models had
been developed to predict SAP and dropout, however, the sole reliance on
predictive accuracy for the decision on classifier’s performance in this domain
and the unquenchable search for efficient models left room for the develop-
ment of novel and more efficient prediction models. This paper presents
novel hybrid prediction models that can be used to predict students’ academic
performance and dropout using the Recursive Feature Elimination (RFE)
technique. The models are hybrid of Support Vector Machine (SVM) and Na-
ïve Bayes (NB) algorithms. The evaluation of the models in comparison with
the existing models showed a better and well comparable performance.

Keywords: Students Academic Performance, Support Vector Machine, Na-

ïve Bayes,
Recursive Feature Elimination, Dropout

12
  
Abstracts  
Stage classification of Alzheimer’s disease using
Transfer Learning
Tariq Saeed Mian1[0000-0003-2666-9223]
1
Department of IS, College of Computer Science and Engineering
Taibah University, Madinah Almunwarah, Saudi Arabia
tmian@[Link]

Abstract: In Alzheimer’s disease (AD) the tissues and cells in the

effected brain start to die gradually and as a result the brain start to
shrink in size and the patient starts to suffer from memory loss. It is a
progressive neurological disorder, and a typical AD patient gradually
starts to lose memory, the ability to speak, some limb movements, and
the ability to think clearly. Due to the complex and progressive nature
of the disease it is important to study its progression and to understand
the Mild Cognitive Impairment (MCI) at every stage. This research is
an attempt to use Deep Learning (DL) Algorithms to study the gradual
progression of AD and develop a strategy for the classification of its
stages. This research is an attempt to use advanced ML and DL tech-
niques to advance the classification and prediction accuracy of the
progression of AD, hence, to improve patient care. In previous studies
structural magnetic resonance imaging (MRI) was used to study the
progression of the disease. In this study dataset from Alzheimer’s Dis-
ease Neuroimaging Initiative (ADNI) is used to study the AD pro-
gression using ML and DL algorithms. In this dataset MRI and PET
images of the brains, the patients’ genetic information, test of cogni-
tive ability, CSF and blood biomarkers are used as predictors of AD.
This dataset has four different classes. In this study the progression of
AD is evaluated using three Convolutional Neural Network (CNN)
Architectures with transfer learning. The proposed CNN architecture
in this study is VGG16, ResNet50 and Dense201. The DenseNet201
model outperform and have the accuracy of 0.95%.

Keywords: ADNI, VGG, DenseNet, ResNet and Magnetic Reso-

nance Imaging.

13
  
Abstracts  
Predicting the e-signing likelihood of loan using
machine learning models combining clustering
with supervised learning
Abdellatif Aattouf 1, Said El Kafhali1 [0000-0001-9282-5154] and Youssef Saadi 2
1
Computer, Networks, Modeling and Mobility Laboratory (IR2M),
Faculty of Sciences and Techniques, Hassan First University of Settat,
B.P. 577, 26000-Settat, Morocco
abdelaat@[Link], [Link]@[Link]

2
Information Processing and Decision Support Laboratory,
Faculty of Sciences and Techniques, Sultan Moulay Slimane University,
Beni Mellal, Morocco
[Link]@[Link]

Abstract. Within the context of acquiring a loan via specific plat-

forms, we propose a hybrid algorithm to predict the user’s behavior
over the process of e-signing the intended loan over the application
and the probability of getting to the final stage of acquiring said loan.
Our methodology incorporates a public dataset made available by a
peer-to-peer Lending website and downloaded from Kaggle for stud-
ying and predicting user e-signing loans. This paper consists of ap-
plying numerous machine-learning algorithms and comparing them
on that dataset. Furthermore, we are proposing a new approach which
consists of a hybrid method combining an unsupervised learning al-
gorithm with a supervised one in order to outperform the best one
from the previous applied ones. Our hybrid model; based on compar-
ing the accuracy of prediction between the application of Random
Forest on the sample as a whole and the application of said model on
specific clusters of said sample, thus choosing the method with the
most accurate prediction; has succeeded in surpassing the best global
model on two clusters over sixteen therefore raising the global Key
Performance Indicators (KPIs).
Keywords: E-signing a Loan prediction, Machine Learning, Self-Or-
ganizing Map, Artificial Neural Network, Supervised learning, unsu-
pervised learning

14
  
Abstracts  
Designing an Arabic Google Play Store User
Review Dataset for Detecting App Requirement
Issues
Maha Al-Shamani1, Mohammed Al-Sarem1, Faisal Saeed1, and Wafa'a
Almutairi1
1
College of Computer Science and Engineering, Taibah University, KSA
tu4160205@[Link];msarem@[Link];
fsaeed@[Link]; w-alsari93@[Link]

Abstract. User reviews in app stores are considered very rich mainte-
nance information texts that developers need to know. A huge number
of app reviews contain requirement details, such as bugs or problems,
evaluation of user experience with some features, suggestions for im-
provements, and ideas for new features. Previous literature has illus-
trated different techniques and approaches to reduce the work needed
to analyze and extract valuable content from mobile app reviews.
However, little attention has been paid to analyzing and studying Ar-
abic app reviews. This paper presents a dataset of some reviews under
governmental service applications, which can provide an overview of
the types of requirement issues that users report on the apps. The da-
taset contains around 51,000 user reviews from six applications re-
trieved from the Google Play Store, which were extracted with the
scraping technique. In this dataset, different forms of analysis were
conducted. The potential usefulness of the dataset was discussed for
future research in the field of software development and improvement
tasks. The dataset was made accessible to the public to support the
community in further studies on the detection of app deficiency or
requirement issues from Arabic reviews.

Keywords: Arabic review dataset, Google Play Store, application re-

quirements.

15
  
Abstracts  
A Survey of Machine Learning and Deep Learing:
Applications in Genome Editing
Talal Almutiri1, Faisal Saeed2 and Manar Alassaf3
1
Department of Information Systems, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah,
Saudi Arabia
2
Department of Information Systems, College of Computer Science and Engineering, Taibah University, Medina, Saudi Ara-
bia
3
Department of Computer Science, College of Computer, Qassim University, Buraydah, Saudi Arabia
1
 [Link]@[Link],
2
 fsaeed@[Link]
3
 [Link]@[Link]
Abstract. A genome is a complete DNA segment that contains the ge-
netic information of how the organism cell lives and performs its func-
tions, such as producing proteins. The field of biology that studies the
complete information about the genome is known as genomics. It intro-
duces concepts and techniques about the sequence, functions, evolution,
and mapping of the genome. Many genomics applications in broad areas
improve creatures’ lives via precise diagnosis and effective treatment.
Analyzing the vast amount of genetic information generated from medi-
cal technologies and devices is a complicated process. Therefore, imple-
menting deep learning and machine learning with other methods such as
feature selection and clustering reduces cost and time spent and enhances
prediction accuracy. Scientists can discover hidden patterns in genetic
information by utilizing machine learning models, finding mutations in
gene expression, and selecting precise locations in gene editing. This
study reviewed recent studies that applied machine learning approaches
in genomics and focused explicitly on gene editing. One of the hot topics
for gene editing research is offtarget mutation prediction using Clustered
Regularly Interspaced Short Palindromic Repeats (CRISPR) and
CRISPR-associated protein 9 (CRISPRCas9). Prediction techniques in
gene editing have been developed to determine values based on mis-
matches in the CRISPR-Cas9 guide series. This paper summarizes recent
studies that have employed machine learning and deep learning methods
in different genomics applications and presents a comparative analysis of
the discussed studies.

Keywords: Deep Learning, Gene Editing, Genomics, Machine Learning.

16
  
Abstracts  
Educational Data Mining techniques for detecting
undesirable students’ behaviors and predicting
students’ performance: A comparative study

Imane EL MOURABIT and Said JAI ANDALOUSSI and Noreddine

ABGHOUR

Faculty of Sciences Ain Chock, Hassan II University, 20100 Casablanca Morocco

[Link]@[Link]
[Link]@[Link]
nabghour@[Link]

Abstract. Nowadays, traditional and online educational systems

generate huge amounts of data. Extracting the useful knowledge and
the hidden patterns from this data can help decision makers to im-
prove teaching and learning by detecting undesirable students’ be-
haviors and predicting students’ performance. Educational Data Min-
ing is the application of data mining techniques for extracting valua-
ble information from huge amounts of data collected during teaching
and learning. This paper presents a state of the art of two tasks Edu-
cational Data Mining that have gained from machine learning and
deep learning, namely detecting undesirable students’ behaviors and
predicting students’ performance.

Keywords: educational data mining, machine learning, deep learn-

ing, educational

17
  
Abstracts  
COVID-19: Analysis and measurement of the
influence of the tweets to help the public health
sector to fight Coronavirus

1 2 3
Abdessamad Essaidi , Dounia Zaidouni , and Mostafa Bellafkih

RAI2S, INPT, Rabat, Morocco

1
[Link]@[Link], [Link]@[Link], and
3
mbellafkih@[Link]

Abstract. Dissemination of information related to coronavirus in all

its key forms, including on Twitter, analysis of tweets mentioning
COVID-19, public attitudes towards COVID-19, and social distanc-
ing are critical to the fight against the Coronavirus. In this work, we
propose a method to measure the influence based on Retweet, and we
highlight the importance of retweeting as a principal method of prop-
agation of information to reduce the spread of Coronavirus in an im-
portant time. Finally, we use several comparisons to other ranking
systems to show highly influencer and use these influencers to help
the public health sector to fight Coronavirus.

Keywords: COVID-19, Coronavirus, Twitter, Public health, Influ-

ence, Retweet, Followers

18
  
Abstracts  
Data Analytics Model for Home Improvement
Store
Lim Jia Yeng, Mohd Norshahriel Abd Rani, and Nabilah Filzah Mohd Radzuan
Centre for Emerging Technologies in Computing (CETC), Faculty of Information
Technology, INTI International University, Negeri Sembilan, Malaysia
limjyenh1999@[Link],
[Link]@[Link],
[Link]@[Link]

Abstract. The Home improvement store is one of the popular retail stores
that sells different categories of products for households such as decora-
tions, accessories, and furnishing. The competition among the home im-
provement stores is very intense where the high business flow has led to a
large and overwhelming amount of business data. Thus, poor handling of
business data brings some negative impacts on the business. They cannot
make a wise business decision if the data is not being utilized to transform
it into useful information. The purpose of this study is to design data ana-
lytics model for the home improvement store in order to solve these issues.
The selected dataset that includes the sales information of the home im-
provement store will be evaluated and pre-processed before implementing
it for analytics. The historical and current business sales will be analysed to
find out the customer preferences toward the products through descriptive
analysis especially for customer segmentation using k-mean algorithm. Pre-
dictive analytics can also be applied to predict the future business sales or
trends by building a regression model using linear regression algorithm.
The set of standard and accurate business rules will also be designed for
this study based on the dataset. The analysis results will be interpreted and
evaluated to identify the model accuracy before conclusions can be made.
The analytics results will be visualized in a customizable web-based dash-
board so that the clients can apprehend it which the shop owner can make
efficient business strategy to increase revenue.

Keywords: Predictive Analytics, Descriptive Analytics, Business

Analytics.

19
  
Abstracts  
Intelligent Models for Mining Social Media Data

Sanaa Elyassami1, Sultan Albloushi1, MohamedAmmar Alnuaimi1, Omran

Alhosani1, Hassan Al Ali1, Khalid Almarashda1,
1
Abu Dhabi Polytechnic, Abu Dhabi, UAE
[Link]@[Link]

Abstract. Social media produces enormous amounts of unutilized

data. Through machine learning and text analytics techniques, organ-
izations and individuals alike can use such data to extract a great deal
of information. The type of information can be used in a plethora of
ways; With techniques such as sentiment analysis allowing organiza-
tions to determine public reception to their products or finding trendy
topics through the application of topic extraction. This paper outlines
the process taken to develop a machine learning model that analyzes
social media posts and extracts the post’s sentiment and topic. A Mul-
tinomial Naïve Bayes model and a random forest model were trained
for each objective to gauge which learning algorithm performs better
with accuracy and F-Score being performance metrics. The results
showcased that even with relatively simple implementations, the anal-
ysis results on unseen social media posts produced promising accu-
racy results with short training time, with the Multinomial Naïve
Bayes model outperforming the random forest model.

Keywords: Social Media, Data Mining, Machine Learning, Random

Forest, Naïve Bayes, Sentiment Analysis, Topic Extraction

20
  
Abstracts  

Cloud computing & Networking

21
  
Abstracts  
Routing and charging of electric vehicles: survey
Meryem ABID1 and Mohamed TABAA1
1
Pluridisciplinary Laboratory of Research and Innovation (LPRI), EMSI
Casablanca, Morocco

[Link]@[Link]

Abstract. Automotive industry is one the most active industries all

over the world. It knows a fierce competition between manufacturers
who aim to design the perfect vehicle, thus, the shift towards electric
vehicles. Known for their emission-free, electric vehicles provide an
outstanding alternative for combustion vehicles which have been ac-
cused of contributing in the aggravation of greenhouse effect. What
makes electric vehicles stand out, besides having little to no impact
on the environment, is their low consumption and the silent, smooth
driving experience they offer. However, there are two of the most
controversial concepts standing in the way of fully adopting electric
vehicles. First, these vehicles are expected to be intelligent, therefore
they are equipped with an advanced computer on board that allows
them to communicate with their surroundings and make decisions
such as choosing a specific path to reach their destination. With that
in mind, the aim is to spend the least time possible to travel around.
Second, to this day, and despite the technological evolution, storing
electricity and retrieving all of it later on for use is still an issue. Thus,
this paper provides an overview of some of the solutions proposed by
recent works regarding the two aforementioned issues, taking into
consideration constraints like time, consumption, and available re-
sources. Different algorithms and approaches regarding Vehicle
Routing Problem will be discussed. Then, the charging of electric ve-
hicles will be analyzed while taking into consideration the battery life-
time and the imposed constraints in each case.

Keywords: electric vehicle, optimal path, VRP, charging, battery

lifetime, energy consumption, time window.

22
  
Abstracts  
A Parallel Genetic Algorithm for solving the
Vehicle Routing Problem with Drone medication
delivery

Mustapha OUISS1, Abdelaziz ETTAOUFIK1, Abdelaziz MARZAK1,

Abderrahim TRAGHA1,
1
Faculty of Sciences Ben M’sik, Hassan II University Casablanca, Kingdom of
Morocco

[Link]-etu@[Link]

Abstract. Drones routing, also known as unmanned aerial vehicles

routing, has become a highly active area of research. However, the
healthcare domain didn't raise enough attention in comparison with
other research domain as parcel delivery, or surveillance, and other
domains.
In this article, we propose to solve a routing problem with drone to
deliver medication, using a parallel genetic algorithm to reach an op-
timal solution in a reduced time execution. The algorithm aims to find
the optimal routes of drones. We consider the problem as a multiple
depots vehicle routing problem with drone. The approach is to create
clusters of patients, as depots are the centroids. Each cluster is opti-
mized with an instance of the genetic algorithm executed using mul-
tiple cores of the processor.

Keywords: Vehicle Routing Problem, Multiple Depots, Drones,

Healthcare, Genetic Algorithm, Delivery, Parallel Computing.

23
  
Abstracts  
Review on Blockchain & Access Control Systems

AMALLAH Mohamed Anass *, ABGHOUR Noreddine, MOUSSAID

Khalid, EL OMRI Amina, RIDA Mohamed

LIMSAD Labs, Faculty of Sciences, Hassan II University of Casablanca, 20100,

Casablanca, Morocco
*
[Link]@[Link]

Abstract. Blockchain is an emerging technology that can demon-

strate its usefulness through its characteristics (decentralization, dis-
tribution, peer-to-peer transmission, immutability, transparency …)
in different area such as security in general, data management and
access control in particular. Access Control Systems (ACS) play a
considerable role in the process of granting rights to resources access
and assets management. However, existing distributed ledger-based
access control still have challenging privacy issues, such fine-grained
access control, confidentiality, integrity and availability. The goal of
this paper is to introduce blockchain and briefly access control sys-
tems and to present some works realized in the realm of blockchain
and access control systems. A discussion is engaged about a compar-
ative table which concerns some results previously introduced. Inter-
ests are expressed on access control in healthcare emergency cases
and ways are suggested for more deeply research. In the same section,
future works are also announced for further investigations.

Keywords: blockchain, access control systems, emergency cases.

24
  
Abstracts  
Adoption of Docker Containers as an Infrastruc-
ture for Deploying Software Applications: a review
W.M.C.J.T. Kithulwatta1, Wiraj Udara Wickramaarachchi2,3 , K.P.N. Jayasena4 , B.T.G.S. Ku-
mara4 and R.M.K.T. Rathnayaka5
1
Faculty of Graduate Studies, Sabaragamuwa University of Sri Lanka, Belihuloya, Sri Lanka
chiranthajtk@[Link]
2
Department of Computing, Faculty of Applied Sciences, Rajarata University of Sri Lanka,
Mihintale, Sri Lanka
3
School of Computer Science & Technology, Wuhan University of Technology, Wuhan,
Hubei Province, China
wirajudara@[Link]
4
Department of Computing & Information Systems, Faculty of Applied Sciences,
Sabaragamuwa University of Sri Lanka, Belihuloya, Sri Lanka
{kumara, pubudu}@[Link]
5
Department of Physical Sciences & Technology, Faculty of Applied Sciences, Saba-
ragamuwa University of Sri Lanka, Belihuloya, Sri Lanka
kapilar@[Link]
Abstract. Containerized infrastructure and containerized software applications
have become one of the most popular topics in the field of Information Technol-
ogy today. Containerization has been identified as an alternative for virtualization.
However, due to the challenges and threats of virtualization, containerization has
become attractive among many users and organizations; especially, Docker has
become the most popular to deploy software applications or services in the last
few years among all container vendors’. The current study is mainly focused on
the features of containers and differences of Docker containers, which are ad-
dressed in the modern literature. The main objective of this research study is to
understand how the Docker containers contribute as value-added infrastructure for
software application deployments based on reputed twenty-six literatures, which
were summarized which are available in public. The study emphasizes the theo-
retical and practical relevance of Docker containers to make a more featured in-
frastructure that launched a comprehensive investigation, in comparison to the re-
spective virtualizations. According to the literatures, the Docker containers have
achieved a grand market share.
Keywords: Containers, Docker, Infrastructure, Virtualization.

25
  
Abstracts  

Study of the impact of the profoundness of tun-

nel in the context of transition from IPv4 to IPv6
at the local, regional, and national levels.

Dounia EL IDRISSI, Najib ELKAMOUN ,and Rachid HILAL

Chouaib Doukkali University, Route Ben Maachou, 24000, Morocco

[Link]@[Link]

Abstract. The IPv4 (Internet Protocol version 4) are about to be ex-

hausted and the next generation of Internet Protocol version 6 (IPv6)
is gradually deployed over the Internet. Due to the incompatibility
between these two versions, a phase of coexistence between these two
versions is necessary. For this, various transition mechanisms have
been developed. The IPv4 to IPv6 transition based on the tunneling
mechanism consists in creating a tunnel between two remote IPv6
networks, which communicate with each other through an IPv4 infra-
structure. Whereas, the tunnel profoundness presented by the number
of routers crossed by the tunnel degrades the performance of the net-
work due to processing at the router level. In this paper, we present a
study of the impact of tunnel profoundness used in the transition from
IPv4 to IPv6, at the local, regional, and national levels, for the most
used routing protocols. The results indicate that profoundness of tun-
nel degrades network performance by up to 15%, 50%, and 65%, re-
spectively at local, regional, and national levels. The best routing pro-
tocol combination is OSPF and RIPng, which offers a maximum im-
provement of 25%.

Keywords: IPv4, IPv6, Transition, Tunnel, OSPFv2, OSPFv3, RIP,

RIPng, and EIGRP.
26
  
Abstracts  

Artificial Intelligence II

27
  
Abstracts  
A Face Mask Detection System Based on Deep
Learning Convolutional Neural Networks
1
Pierre Stanislas Birame Ndong , Wilfried
2;5 2
Yves Hamilton Adoni , Tarik Nahhal ,
1 3;4
Charles Mberi Kimpolo , Moez Krichen ,
2 2
Elbyed Abdeltif , Ismail Assayad , and
1
Franck Kalala Mutombo
1
  African Institute of Mathematical Sciences (AIMS), Mbour, Sene-
gal
2
  Hassan II University of Casablanca, Science Faculty, Casablanca,
Morocco
3
4
  CSIT, Albaha University, Albaha, Saudi Arabia
5
  ReDCAD Laboratory, University of Sfax, Tunisia
  International University of Casablanca

Abstract. The alarming situation caused by COVID-19 has prompted

the Arti cial Intelligence community to think about using new tools in
order to contribute to the ght against COVID-19. Computer vision is
a sub-discipline of AI that is blossoming since 2010s and helping in
diverse elds like robotics, self-driving cars, health care by handling
very com-plex non linear problems. In the wake of this it’s contribu-
tion to the ght against COVID-19 has to be taken into account. In fact
only a few pa-pers have been published recently in that regard despite
the seriousness of the situation. This is what fueled our desire to make
a contribution. This paper is intended to elaborate very powerful and
robust mask de-tection model enabling the detection of not only indi-
viduals wearing the recommended masks from those who don’t but
the detection of a third category of people not wearing it appropriately
is also taken into account by our model.

Keywords: Covid-19 Deep learning Convolutional Neural Networks

Object Detection Facial Recognition Region Proposals Real-Time
Ob-ject Detection OpenCV Ca e YOLOv3.

28
  
Abstracts  
Deep learning for COVID-19 cases based XCR
and chest CT images
Anmar Abuhamdah1, Ghaith M. Jaradat2, and Mutasem Alsmadi3
1
Department of Management Information System, College of Business
Administration, Taibah University, Saudi Arabia
aabuhamdah@[Link]
2
Department of Computer Science, Faculty of Computer Science and Informatics,
Amman Arab University, Jordan
[Link]@[Link]
3
Department of Management Information Systems, College of Applied Studies and
Community Service, Imam Abdurrahman Bin Faisal University, Saudi Arabia
mksalsmadi@[Link]

Abstract. The ferocity of Coronavirus Disease COVID-19 made CT

scans urgently needed as a practical and effective alternative. Emer-
gency departments may favor this alternative due it is mobility, accu-
racy, speed, inexpensive, and less radiologist training. Hence, it is re-
quired to develop an automated and accurate detection using chest CT
scans. Therefore, this study employs a hybrid predictive model. This
predictive model comprises a deep convolutional neural network us-
ing WekaDeeplearning4j. It extracts chest X-ray and CT imaging fea-
tures from pneumonia or other lung diseases training cases. Then, it
further refines those features with COVID-19 cases to train selected
imaging features capable of classifying three cases. They are classi-
fied as positive, pneumonia, and negative. This study demonstrates
capability and effectiveness of the classifier, for detecting positive
cases through an experimental performance. This model is tested on
a number of reliable and up-to-date datasets such as John Hopkins
University, ieee8023 and Kaggle. Trained models of the classifier
showed promising detection results in terms of accuracy and speed.

Keywords: COVID-19 detection, XCR and chest CT imaging, Fea-

tures Extraction, Classification, [Link].

29
  
Abstracts  
Performance comparison of K-Nearest Neighbor,
Random Forest, and Multiple Linear Regression
to predict photovoltaic panels' power output

CHAHBOUN Souhaila1 and MAAROUFI Mohamed1

1
Mohammed V University of Rabat, Mohammadia School of Engineering, Ibnsina
B.P. 765 Agdal Rabat 10090, Morocco
souhaila.chahboun93@[Link]
maaroufi@[Link]

Abstract. Recently with the emergence of Industry 4.0 and infor-

mation technologies, Machine learning techniques within artificial in-
telligence (AI) have become very popular and widely used in many
fields such as renewable energies (solar energy…). In fact, although
Machine learning technologies have existed since the 1960s, it is only
in the last years that they have been really exploited and have attracted
many authors particularly with the advent of new algorithms that en-
able to benefit from the analysis and prediction capabilities of Ma-
chine learning as well as Data Lakes that are continuously produced
in real time by sensors and Internet of Things (IoT) activators. Thus,
one of the core objectives of this paper is to assess the performance
of classical or traditional Machine learning methods namely Multiple
linear regression and advanced ones such as Random forest and K-
Nearest Neighbor for the short-term prediction done for one hour of
the power generated by photovoltaic panels using multiple forecasters
and identify the model that offers the best accuracy by using the most
popular performance metrics : Root Mean Square Error, R-squared
and Mean absolute error.

Keywords: Artificial intelligence, Machine learning, Photovoltaic

output power

30
  
Abstracts  
Voting Classifier-based Intrusion Detection for
IoT
Networks

Muhammad Almas Khan1, Muazzam A Khan1, Shahid Latif2, Awais Aziz Shah3,
Mujeeb Ur Rehman4, Wadii Boulila5;6, Maha Driss5;6, and Jawad Ahmad7
1
Quaid-i-Azam University, Islamabad, Pakistan
2
Fudan University, Shanghai, China
3
Polytechnic University of Bari, Italy
4
Riphah International University, Islamabad, Pakistan
5
University of Manouba, Manouba, Tunisia
6
Taibah University, Medina, Saudi Arabia
7
Edinburgh Napier University, Edinburgh, UK

Abstract. Internet of Things (IoT) is transforming human lives by

paving the way for the management of physical devices on the edge.
These interconnected IoT objects share data for remote accessibility
and can be vulnerable to open attacks and illegal access. Intrusion de-
tection methods are commonly used for the detection of such kinds of
attacks but with these methods, the performance/accuracy is not opti-
mal. This work introduces a novel intrusion detection approach based
on an ensemble-based voting classifier that combines multiple tradi-
tional classifiers as a base learner and gives the vote to the predictions
of the traditional classifier in order to get the final prediction. To test
the effectiveness of the proposed approach, experiments are per-
formed on a set of seven different IoT devices and tested for binary
attack classification and multi-class attack classification. The results
illustrate prominent accuracies on Global Positioning System (GPS)
sensors and weather sensors to 96% and 97% and for other machine
lear ing algorithms to 85% and 87%, respectively. Furthermore, com-
parison with other traditional machine learning methods validates the
superiority of the proposed algorithm.

Keywords: IoT; Intrusion Detection; Machine Learning; Classifica-

tion; Voting Classifier

31
  
Abstracts  
Glazing's Potential Energy Savings for Green
Buildings

Badr Chegari1,2,4[0000-0002-5664-6543], Mohamed Tabaa2, Emmanuel Simeu4, Fouad Moutaouakkil1,2,

Hicham Medromi1,2
1
Research Foundation for Development and Innovation in Science and Engineering
(FRDISI), Casablanca, Morocco.
2
Engineering research laboratory (LRI), System Architecture Team (EAS), National and high
school of electricity and mechanic (ENSEM), Hassan II University, Casablanca, Morocco.
3
Pluridisciplinary Research and Innovation Laboratory (LPRI), Moroccan School of
Engineering Sciences (EMSI), Casablanca, Morocco.
4
Univ. Grenoble Alpes, CNRS, Grenoble INP, TIMA, 38000 Grenoble, France*.
* Institute of Engineering Univ. Grenoble Alpes.

Abstract. Energy savings in the building sector are important for achieving
sustainable development. Actually, the building sector in Morocco is poorly
adapted to the country's climatic conditions. This contributes to an increase in
electricity demand of 7% per year. The Department of Energy is interested in
introducing sustainable building into major construction programs in Mo-
rocco. Especially, since there is a strong demand for new buildings, for which
the general objective is to build them more energy-efficient than in current
practice. To achieve this objective, the energy demand of the building must be
reduced as much as possible without compromising living comfort. This study
presents the potential advantages of the glazed shutter regarding the energy
performance of a typical Moroccan ground floor + first floor (GFFF) building
located in Marrakech city. The results were obtained mainly through dynamic
thermal simulations using DesignBuilder software. These show that the glazed
shutter of the building has a significant impact on its energy performance,
which reduces the heating and cooling demand by 22% for the studied climate.

Keywords: Energy efficiency, Green Building, Glazing, Thermos-

physical parameters, Dynamic thermal simulation.

32
  
Abstracts  
Regenerating a Graphical User Interface Using
Deep Learning
Fatima CHEDDI1, Ahmed TAHIRI2 and Mohammed SERRHINI3
Laboratory of Arithmetic Scientific Computing and Applications, Mohammed I
University in OUJDA, Morocco
1
[Link]@[Link],2Tahiriahmed02@[Link]
3
serrhini@[Link]

Abstract. Transforming a conceptual drawing of a Graphical User

Interface UI into computer code is a long and complex process, which
a time-consuming work and re- quires an expertise developer. To
overcome this problem, we propose an automated system using com-
puter vision and optical character recognition (OCR) techniques. This
paper proposes a novel hybrid approach using a Convolutional Neu-
ral Networks (CNNs), which is composed of Alex Net, Single Shot
Detectors (SSDs) and Recurrent Neural Networks, in order to detect
User Interface elements such as images retrieval, texts, button, links,
and lists in the input sketch. The output from the network is a file
representing a list of UI elements detected along with their properties.
We evaluate our Model on two public datasets including Sketched
elements, and Iam datasets. Our Experimental results demonstrate
that our hybrid approach achieves over 90% of accuracy.

Keywords: Convolutional neural network, SSD, Classification,

Handwriting recognition, OCR.

33
  
Abstracts  

Information Security

34
  
Abstracts  

A Review of Blockchain and the Benefits for

Digital Marketing-related Applications of
Blockchain Integration

Mouhssine Abirou*, Noureddine Abghour

IMSAD Lab, Dept of Mathematics and Computer Science, Faculty of Science Ain-
Chok, Hassan II University of Casablanca
Casablanca, Morocco
[Link]-etu@[Link]

Abstract. Blockchain can be defined as a network and at the same

time a database. It’s a distributed and secured architecture based on
advanced cryptographic system. Through the theoretical drive, this
article overviews the main notions related to blockchain’s character-
istics, its activity’s fields and whose development has progressed very
quickly from digital currency, digital finance, and digital society to
digital intelligence. Nowadays, companies should capitalize emerg-
ing Martech solutions and shift their marketing strategies to achieve
their targets. Therefore, digital marketing is a modernist aspect of
blockchain technology. A lot of key issues may be addressed if block-
chain and digital marketing are mixed.

Keywords: Blockchain, Digital Marketing, Business, Smart Con-

tract, Customer, Trust, Privacy.

35
  
Abstracts  

Secure Remote Workplace 4EM model

Cakula Sarma

Vidzeme University of Applied Sciences, Valmiera LV 4201, Latvia

[Link]@[Link]

Abstract. The global pandemic, which began in 2020, ushered in a

significant time in the field of information technology, when the per-
formance of work duties, the receipt of services and mutual commu-
nication became most directly dependent on information and commu-
nication technologies. When using information and communication
technologies, communication is diverse. In addition, not all commu-
nication is verbal. This new digital work environment poses new
problems and challenges for business leaders. The main goal of this
paper is to create a secure remote workplace model that includes fac-
tors for secure communication in forced distancing conditions, as well
as to provide recommendations for companies. The model has been
developed using the 4EM method, describing the factors involved and
the attributes to be measured. Business process analysis and automa-
tion includes identification and rationalization of processes, execution
of business processes with minimal human involvement. The devel-
oped model provides tools for better understanding the relationship
between all the used data sources and storages in the context of the
expected results therefore the data could be shared effectively across
the different sources.

Keywords: Communication technologies, Communication security,

4EM method.

36
  
Abstracts  
Review in Authentication for Mobile Ad hoc
Network

Khalfaoui Hafida, Farchane Abderrazak, and Safi Said

Sultan Moulay Slimane University, LIMATI Laboratory,
Beni Mellal, Morocco
hafidakhalfaoui1996@[Link], [Link]@[Link],
[Link]@[Link]

Abstract. As far back as the 90s, mobile Ad hoc networks have suc-
cess- fully gained the scientific community interest because of the
ease of their deployment and administration in many domains such as
military, collaborative work, crisis conditions and environmental ap-
plications. This network technology provides opportunities for com-
panies to reduce soft- ware operating costs at anytime and anywhere.
However, authentication of nodes is a challenging task due to its open
environment and resource constraint infrastructure. For this purpose,
we will present in this paper a survey about some diff erent techniques
of authentication of Mobile Ad hoc Networks existing in the litera-
ture.

Keywords: Mobile Ad hoc Network · Authentication · Security

37
  
Abstracts  
A Panoramic Evaluation of Machine Learning and
Deep Learning Aided Ransomware Detection
Tools Using a Hybrid Cluster of Rich Smartphone
Traits
Hiba Zuhair1
1
Department of Systems Engineering, College of Information Engineering,
Al-Nahrain University, Baghdad, Iraq.
[Link].pcs2013@[Link],
[Link]@[Link],
[Link].pcs2013@[Link]

Abstract. Today, users adopt smartphones as their primary, mobile,

and connective computing platforms for communication, and big data
storage. However, users are strongly threatened by the time-variant
ransomware versions which increase their duplicity, excess, decep-
tions, and then damage to defeat the existing artificially intelligent
ransomware detection tools due to their limits at halting ransomware
versions and mitigating their running propagation online. Hence, this
paper spots the light on an artificially intelligent solution that applies
a rich cluster of hybrid traits and elaborates more robust and effica-
cious ransomware detection tools in smartphones. To do so, 24 static
and dynamic traits with six machine learners and three deep learners
are proven experimentally to affirm their discrimination power, their
efficacy and efficiency, and their robustness versus time-variant ran-
somware versions of 27 various android ransomware families. Ex-
perimental outcomes approve that the trait cutoffs could enrich the
decisive classification in the near future.

Keywords: ransomware versions; crypto-ransomware; locker ran-

somware; smartphone security; android platform; file system analy-
sis; machine learning; deep learning.

38
  
Abstracts  
A Big Data driven model for secured systems:
A Blockchain-based architecture for Cloud
Computing and IoT security

Aristide NDAYIKENGURUKIYE1, Abderrahmane EZ-ZAHOUT1 and

Fouzia OMARY1
1
Mohammed V University, Rabat 4 Ibn Batouta Avenue, Morocco
Aristide_ndayikengurukiye@[Link],
[Link]@[Link] and omary@[Link]

Abstract. With the advent of new networked elements such as smart cities,
smart homes, the remarkable development of the Internet of Things (IoT) has
been driven, which has enabled the use of small devices to manage and per-
form complex tasks. Around these devices has come the incessant collection
of a mass of unstructured data that sometimes needs to be analyzed and stored
in the Cloud computing, which has enabled the development of multiple ap-
plications for their proper functioning. The biggest challenge of information
systems, especially when using IoT and Cloud computing, remains the secu-
rity of the data and the permanent availability of the information. Blockchain
is one of the promising solutions applied to the various security problems
today. In this paper, we propose a new architecture based on Blockchain tech-
nology to ensure the security of IoT and Cloud Computing. A literature re-
view on the use of Blockchain for IoT security has also been presented.

Keywords: IoT, Cloud Computing, Big Data, Blockchain and Security.

39
  
Abstracts  
Hybrid Filter and Wrapper Feature
Selection Method for Enhancing Detection
Process of Phishing Websites

Wafa'a Almutairi*, Faisal Saeed*, Mohammed Al-Sarem, and

Maha Alshamani

College of Computer Science and Engineering, Taibah University,

Medina, KSA
*Corresponding authors:
w-alsari93@[Link];
fsaeed@[Link]

Abstract. In this age of data and digital technology, phishing is con-

sidered a major problem due to its huge impact on the financial and online
retailing sectors. All commercial activities on the Internet are the target
of cybercriminals, so it is important to take security measures to protect
assets. Preventing a phishing attack before it happens is one of the initial
stages of creating secure cyberspace. This paper proposes a feature selec-
tion method that investigates several combinations of filter and wrapper
methods to minimize conflict in feature selection results and uses multiple
classification models to provide better results on phishing website detec-
tion. The experimental results showed that the detection rate of phishing
websites was improved when using the feature selection methods.

Keywords: Feature selection, phishing websites, detection,

classification, cybercrime.

40
  
Abstracts  

Internet of Things

41
  
Abstracts  
Optimization of Wind Farms Based on the
Selection of Types of Wind Turbine

Abdelouahad Bellat1, Khalifa Mansouri1, Abdelhadi Raihani1

1
SSDIA Laboratory, Hassan II University of Casablanca, ENSET of Moham-
media, 28830, Morocco
bellatabdelouahad@[Link],
khmansouri@[Link], abraihani@[Link]

Abstract. Finding a suitable design for the development of a wind

farm (WF) is a complex task in a wind project, hence the optimization
methods that become important to deal with this complexity, espe-
cially with the wide evolution of areas WF. The objective of this study
is the optimization of wind farms according to the selection of wind
turbines and the choice of the appropriate objective function. The in-
formation collected on a given wind farm makes it possible to choose
the wake model and the types of wind turbines to be installed. The
Horns Rev offshore wind farm was chosen to test the validation of
our approach. The application of a genetic algorithm able to select
suitable wind turbines was tested for 3 well-defined objective func-
tions. The results of this algorithm show a significant improvement in
energy production and capacity factor.

Keywords: Wind Farm, Optimization, selection, types of wind tur-

bines, genetic algorithm.

42
  
Abstracts  
Heuristic methods for the container stacking
problems: A review

Farah AL KADERI1, Rim KOULALI2, Mohamed RIDA3

LIMSAD Laboratory, Mathematics and computing Departement,

Faculty of Sciences Ain Chock, Hassan II University

Casablanca, Morocco
1 2
[Link]@[Link], [Link]@[Link],
3
[Link]@[Link]

Abstract. Container stacking is a crucial task of the maritime con-

tainer terminal management process. Any rearrangement of the con-
tainers in storage areas requires a large amount of effort, time, and
cost. Hence, the container stacking problem attracts the attention of
researchers and many works have been proposed to improve the con-
tainer terminals services. In this context, the objective of this paper is
to give a review of the state-of-art and make a comparison between
the proposed solutions. We will present and analyze each proposed
solution and present our perspectives regarding an optimized solution
to the container stacking problem.

Keywords: Container stacking problem, pre-marshalling problem,

relocation problem, maritime container terminals, review.

43
  
Abstracts  
A Smart Power Grid architecture for micro-
neighborhood based on Green IoT

Asmae CHAKIR1,2,3, Mohamed TABAA1, Fouad MOUTAOUAKKIL2,3,

Hicham MEDROMI2,3, Karim ALAMI1
1
Pluridisciplinary Laboratory of Research and Innovation (LPRI), Mo-
roccan School of Engineering Sciences (EMSI), Casablanca, Morocco
2
Research Foundation for Development and Innovation in Sciences and
Engineering (FRDISI), 16469 Casablanca, Morocco
3
Engineering Research Laboratory (LRI), System Architecture Team (EAS),
National High School of Electricity and Mechanic (ENSEM), Hassan II University
of Casablanca, 8118 Casablanca, Morocco
[Link]@[Link]

Abstract. The 21st century and its confrontation with all its present
constraints, notably: the disadvantages of conventional energy pro-
duction, the price of fuel and the rapid urban development, put the
power sector in a challenging situation especially as consumer satis-
faction has become a priority. Therefore, we propose in this paper a
new intelligent structure based on the Green Internet of Things that
facilitates and helps to promote the integration of renewable energy
in the electricity production sector, and supports the emergence of the
notion of producer-consumer units. This architecture is essentially
based on the communication between sustainable households, each
equipped with its own alternative production to reduce the demand on
the utility grid. Each household is considered as a HRES or a distrib-
uted energy point in a micro-grid that works in collaboration to keep
an energetic balance within the grid. Simulations made on
MATLAB/SIMULINK following the architecture and the communi-
cation & energy sharing strategy have shown its efficiency in manag-
ing successive overloads and optimizing home to micro-grid and mi-
cro-grid to home interactions.

44
  
Abstracts  
Keywords: Renewable energy, distributed generation, micro-grid,
peer-to-peer communication, hybrid renewable energy system, sus-
tainable home system, smart grid.

Study of LoRaWAN Performance Under a Noisy

Industrial Channel

Nisrine Bahri1,3, Safa Saadaoui2, Mohamed Tabaa2, Mohamed

Sadik3 and Hicham Medromi1

1
Foundation for Research, Development and Innovation in En-
gineering Sciences, Casablanca, Morocco
2
Pluridisciplinary Laboratory of Research and Innovation
(LPRI), EMSI Casablanca, Morocco
3
RTSE Research team, Laboratory of Research Engineering,
ENSEM, Hassan II university, Casablanca, Morocco
nissrinebahri@[Link]

Abstract. In the latest years, LPWAN has started to gain attention from
researchers to help surmount some of the challenges faced by conven-
tional wireless networks by allowing Industrial Internet of Things (IIoT)
devices to operate efficiently and robustly over long distances, making
them ideally suited for industrial environments. we aim to simulate Lora’s
performances by first an experimental analysis of its physical layer. After
that an informative mathematical examination of the important parame-
ters, we evaluate the performance of LoRa in additive white Gaussian
noise (AWGN) and the industrial noisy channel, using a numerical model
in MATLAB, to determine its communication capabilities under negative
signal-to-noise ratio (SNR). In this paper, we will highlight the perfor-
mance of this recent wireless protocol designed for the Internet of Things
as well as the Industrial IoT, and the impact of impulsive noise, modeled
by the Markov Process, to evaluate the robustness of a LoRa system under
noisy channel.

Keywords: LPWAN, Lora, Wireless Networks, Performances, AWGN,

MATLAB, SNR, Industrial Noise, impulsive noise, Markov process.
45
  
Abstracts  
Weight Initialization Techniques for Deep
Learning Algorithms in Remote Sensing: Recent
Trends and Future Perspectives

Wadii Boulila1,2, Maha Driss1,2, Eman Alshanqiti2, Mohamed Al-Sarem2,

Faisal Saeed2, Moez Krichen3,4
1
RIADI Laboratory, National School of Computer Sciences, University of
Manouba, Tunisia
2
IS Department, College of Computer Science and Engineering, Taibah University,
KSA
3
CS Department, Faculty of CSIT, Al-Baha University, KSA
4
ReDCAD Laboratory, University of Sfax, Tunisia

Abstract. During the last decade, several research works have fo-
cused on providing novel deep learning methods in many application
fields. However, few of them have investigated the weight initializa-
tion process for deep learning, although its importance is revealed in
improving deep learning performance. This can be justified by the
technical difficulties in proposing new techniques for this promising
research field. In this paper, a survey related to weight initialization
techniques for deep algorithms in remote sensing is conducted. This
survey will help practitioners to drive further research in this promis-
ing field. To the best of our knowledge, this paper constitutes the first
survey focusing on weight initialization for deep learning models.

Keywords: Review, deep learning, weight initialization, remote sens-

ing.

46
  
Abstracts  
Toward E-Business during Covid-19: The Business
Case of Saudi Arabia

Eman Bakr Alshanqiti1, Alzahraa Mohammed Alsharif 1, Maha Abdullah

Alshamani 1 and Abdelrahman Elsharif Karrar 1
1
College of Computer Science and Engineering, Taibah University, Saudi Arabia
Amon_sh@[Link];
gorbah@[Link];tbr2012@[Link];
[Link]@[Link]

Abstract. Today, E-business has become an integral part of our daily

life. E-business has evolved into one of the main pillars of the retail
sector and influences companies' decisions worldwide. During the
home quarantine period and the emergence of the Coronavirus in the
world, the world's lifestyle differed. Due to government controls on
closing all markets and shops, which negatively affected business dur-
ing this period. In this research, researchers will discuss how E-busi-
ness was a business continuity plan for businesses during and after
the home quarantine period. In order to study this situation, a ques-
tionnaire is conducted, which was taken by 116 users to demonstrate
their experience of E-business during the quarantine and what is there
criticism of the negatives found, and how E-business could be im-
proved to be the suitable business continuity plan during any crises.

Keywords: E-Business, Business Continuity plan , Recovery Plan,

Saudi Business, Quarantine.

47
  
Abstracts  
Optimization of SNGDBF Decoding for LDPC
Codes parameters based on Taguchi method

Hajar El ouakili1 and Rachid El Alami1, Younes Mehdaoui2, Driss

Chenouni2 and Zakia Lakhliai2
1
Faculty of science,usmba Fez, Morooco
2
ENS,usmba Fez,Morocco

[Link]@[Link]
[Link]@[Link]
[Link]@[Link]
d_chenouni@[Link]
[Link]@[Link]

Abstract. The aim of this paper is to present the Taguchi optimization

method in order to select the optimal parameters of Single Noisy Gra-
dient descent bit flipping (SNGDBF). In this algorithm, three factors
are evaluated: ɳ, iteration numbers and frame. For this purpose, each
of these factors is given three levels and an orthogonal array table is
generated, moreover the signal-to-noise ratio is computed. This meth-
odology applied to the SNGDBF shows that the number of iterations
is a significant factor, in regard to frame and ɳ, which are insignificant
factors until 4db. However, around 5db frame becomes more signifi-
cant. This study also shows that the optimal parameters give good re-
sults compared to previous works in this area. Finally, Taguchi
method is suitable to prove the decision made when it is related to
select parameters of all decoding Low Density Parity Check (LDPC)
algorithms.

Keywords: codes, bit-flipping algorithm, noisy gradient, Taguchi

method, parameters optimization.

48
  
Abstracts  
Deep neural networks hyperparameter
optimization using particle swarm optimization for
detecting frauds transactions
Mohammed Tayibi and Said El Kafhali [0000-0001-9282-5154]
Computer, Networks, Modeling and Mobility Laboratory (IR2M), Faculty of Sciences and
Techniques, Hassan First University of Settat, B.P. 577, 26000-Settat, Morocco
[Link]@[Link], [Link]@[Link]

Abstract. The recent explosions and developments of new technolo-

gies have changed our lives, and that was shown by the quantity of
information shared, posted, and stocked in big companies like Face-
book, Google, Amazon, and so forth. Millions of transactions are
made by cardholders every year to buy online using credit card as a
mobile wallet or for sample payment and that make credit card trans-
action more frequent today. The developments of communication
technologies and E-commerce have made credit cards the most ordi-
nary methods of payment for both online and regular purchases. As a
result, millions of online transactions are subject to various types of
fraud. So security in this system is required to prevent fraudulent
transactions. In this direction, researchers to detect this fraud invent
many approaches. Traditional techniques cannot detect sophisticated
fraudulent. Furthermore, analysis of cardholder behaviors or static
risk management rules of the frauds have never stopped the fraudsters
to commit their crimes. However, artificial intelligence techniques
such as deep learning and machine learning have been able to handle
these issues. This paper proposes an approach to detect fraud transac-
tions by optimizing Deep Neural Networks (DNNs) hyperparameters
using Particle Swarm Optimization (PSO) as optimization methods
and compare them with the grid search (GS) method. The results ob-
tained in terms of precision, accuracy, recall, F1-score, Time, and
Area under the Curve (AUC) have shown that the PSO can generate
better solutions in a short time in comparison with the GS method.

Keywords: Credit card, fraud detection, hyperparameter optimiza-

tion, particle swarm optimization, deep neural networks.
49
  
Abstracts  
EmPRM: A novel Multi-agent model for modeling
and simulating of electromagnetic waves using
Netlogo Platform

Hamid Bezzout 1,* and Hanan El Faylali 1

1
Ibn Tofail University, P.O. Box 242, Kenitra, Morocco
[Link]@[Link]
Wh_elfaylali@[Link]

Abstract. Much has been said about the Electromagnetic waves'

propagation in different media in various studies because of their im-
portance regarding applications for new technologies. To date, there
is, however, no Multi-agent model to simulate this phenomenon. Our
study, a multi-agent model: Electromagnetic Propagation Reflection
Model (EmPRM), is developed to describe the propagation of elec-
tromagnetic waves in air media, the reflection of dielectric media and
their behavior in frequency domain. This model has been validated
through comparison with numerical methods, namely Finite Differ-
ence Time Domain (FDTD). The results for the proposed model are
similar to the results of the previous numerical method reported in
literature. The new model can be extended to other electromagnetic
phenomena.

Keywords: Multi-agent systems, Electromagnetic waves, Netlogo

Platform, Propagation, multi-agent model, dielectric media.

50
  
Abstracts  
Fighting Covid-19 pandemic: what role for digital
technology?

Ouail Ouchetto1,2 and Said Jai-Andaloussi1

1
FSAC, Hassan II University of Casablanca, Casablanca, Morocco
2
FSJES-AC, Hassan II University of Casablanca, Casablanca, Morocco
{[Link],[Link]}@[Link]

Abstract. Coronavirus disease 2019 is a global pandemic with high

transmissibility. All the countries of the world have adopted and ap-
plied different measures to fight against this pandemic, with varying
degrees of success. Countries that have been able to maintain low per
capita death rates from COVID-19 have adopted strategies that in-
clude early surveillance, contact tracing, testing and strict quarantine.
The effective implementation of these strategies requires the support
of new technologies like Artificial Intelligence, Internet of Things,
Deep Learning, Big Data and Blockchain and new mobile apps. This
paper presents the role of these emerging technologies to prevent,
monitor, mitigate and analyze the spread of COVID-19 and it also
presents the challenges related to
the design, development, and use of these technologies. In addition, it
provides some information and suggestions for improving the use of
the technology and producing better solutions to better control and
fight against the COVID-19 pandemic.

Keywords: Covid-19, Artificial intelligence, Big data, Blockchain,

Internet of Things.

51