Prof. Mona Nasr

There are huge data from unstructured text obtained daily from various resources like emails, tweets, social media posts, customer comments, reviews, and reports in many different fields, etc. Unstructured text data can be analyzed to obtain useful information that will be used according to the purpose of the analysis also the domain that the data was obtained from it. Because of the huge amount of the data the human manually analysis of these texts is not possible, so we have to automatic analysis. The topic analysis is the Natural Language Processing (NLP) technology that organizes and understands large collections of text data, by identifying the topics, finding patterns and semantic. There two common approaches for topic analysis, topic modeling, and topic classification each approach has different algorithms to apply that will be discussed.

E-payment is the key function for any e-business as it is rising exponentially in today's business world as e-business grows. E-payments made it easier for people to survive and helped them save a lot of money and time. Using various forms and devices, our payments are more exciting and convenient to press on your mobile phone and pay for your orders. In order to obtain better results in e-business, it must be linked to e-payments. E-payments have many systems and opportunities in the field of e-business, but it is facing many risks and challenges that need to be highlighted in order to find solutions to it. This paper presents an overview study for e-payments opportunities, challenges, and different risks for e-payments especially fraud as it is one of the most critical threats to the e-payments field and it is causing huge losses. Paper also discusses different types of e-payments, benefits, and the future of e-payment.

This paper explains how to detect the 2D pose of multiple people in an image. We use in this paper Part Affinity Fields for Part Association (It is non-parametric representation), Confidence Maps for Part Detection, Multi-Person Parsing using PAFs, Simultaneous Detection and Association, this method achieve high accuracy and performance regardless the number of people in the image. This architecture placed first within the inaugural COCO 2016 key points challenge. Also, this architecture exceeds the previous state-of-the-art result on the MPII Multi-Person benchmark, both in performance and efficiency.

Climate change is a major challenge for agriculture and food security, Climate change may cause sudden reductions in agricultural productivity. Egypt is considered as a one of the worst-weather countries which is affected by the climate change: including high rising temperatures, erratic rainfall, sandstorms and extreme heat. The production of the strategic crops will suffer from significant reduction from 10% to 60% by the middle of the century (2050) due to temperature increases. Intelligent Decision Support System IDSS provides the good solution to minimize the effects of climate change in an intelligent manner and provides the required solutions to successfully penetrate this highly competitive and regulatory challenging environment. These solutions depends on GIS to determine location and its properties, climate prediction model, knowledge base model in agricultural domain as plants, insects and diseases using domain expert knowledge extraction, data mining, machine learning and fuzzy logic to obtain a fast and high accuracy solutions.

Abnormal temperature of human body is a natural extensive indicator of illness. Infrared thermography (IRT) is a fast, non-invasive, non-contact and passive substitution to ordinary medical thermometers for monitoring and observation human body temperature. Aside from, IRT is able to chart body surface heat remotely. Last five decades testified a stationary development in thermal imaging cameras utilization to obtain relations between the thermal physiology and surface temperature. IRT has effectively used in diagnosis and detection of breast cancer, diabetes neuropathy and peripheral vascular disorders. It has been employed to detect issues related to gynecology, dermatology, heart, neonatal physiology, and brain imaging. With the advent of modern infrared cameras, data acquisition and processing techniques, it is now possible to have real time high resolution thermographic images, which is likely to surge further research in this field. The emergent technology known as the Internet of Things (IoT) has guided practitioners, physicians and researchers to design innovative solutions in different environments, particularly in medical and healthcare using smart sensors, computer networks and a remote server. This paper aims to propose IoT-enabled medical system enables diagnostics and detection for several medical anomalies remotely; in real-time and simultaneous depend on combination of IoT and Thermal Infrared imaging techniques. It will detect and diagnostics any abnormal and alert the user through IoT remotely and in real-time.

Diabetes is a inveterate defect and disturbance resulted from metabolic conk out in carbohydrate metabolism thus it has occupied a globally serious health problem. In general, the detection of diabetes in early stages can greatly has significant impact on the diabetic patients treatment in which lead to drive out its relevant side effects. Machine learning is an emerging technology that providing high importance prognosis and a deeper understanding for different clustering of diseases such as diabetes. And because there is a lack of effective analysis tools to discover hidden relationships and trends in data, so Health information technology has emerged as a new technology in health care sector in a short period by utilizing Business Intelligence 'BI' which is a data-driven Decision Support System. In this study, we proposed a high precision diagnostic analysis by using k-means clustering technique. In the first stage, noisy, uncertain and inconsistent data was detected and removed from data set through the preprocessing to prepare date to implement a clustering model. Then, we apply k-means technique on community health diabetes related indicators data set to cluster diabetic patients from healthy one with high accuracy and reliability results.

Abnormal temperature of human body is a natural extensive indicator of illness. Infrared thermography (IRT) is a fast, non-invasive, non-contact and passive substitution to ordinary medical thermometers for monitoring and observation human body temperature. Aside from, IRT is able to chart body surface heat remotely. Last five decades testified a stationary development in thermal imaging cameras utilization to obtain relations between the thermal physiology and surface temperature. IRT has effectively used in diagnosis and detection of breast cancer, diabetes neuropathy and peripheral vascular disorders. It has been employed to detect issues related to gynecology, dermatology, heart, neonatal physiology, and brain imaging. With the advent of modern infrared cameras, data acquisition and processing techniques, it is now possible to have real time high resolution thermographic images, which is likely to surge further research in this field. The emergent technology known as the Internet of Things (IoT) has guided practitioners, physicians and researchers to design innovative solutions in different environments, particularly in medical and healthcare using smart sensors, computer networks and a remote server. This paper aims to propose IoT-enabled medical system enables diagnostics and detection for several medical anomalies remotely; in real-time and simultaneous depend on combination of IoT and Thermal Infrared imaging techniques. It will detect and diagnostics any abnormal and alert the user through IoT remotely and in real-time.

The techniques of data mining are very popular of Diseases. The advancement in health analysis has been improved by technical advances in computation, automation and data mining. Nowadays, data mining is getting used in a vast area .The Nature of the medical field is made with the knowledge wherever there's a spread of data but untapped during a correct. and thus, the foremost serious challenge facing this area is the quality of service provided which suggests to create the diagnose during a correct manner in a timely manner and supply acceptable medications to patients. Thus Health information technology has emerged as a replacement technology within the health care sector in a short amount by utilizing Business Intelligence 'BI' that could be a data-driven Decision Support System. The various techniques of data mining are used and compared during this analysis.

With the new era of Information and Communication Technology; collaborative learning is considered to be an elearning approach where learners are able to socially interact to the others, as well as instructors. In essence, learners work together in order to expand their knowledge of a particular subject or skill. Nowadays, Social Networks could be used as an e-learning platform. One would typically log in and collaborate with other learners on a specific topic using the social network as the common working space. In this paper, a new approach is presented for collaborative learning throughout social networks. The behavior the proposed approach is presented. The proposed collaborative e-learning approach consists of three modules. The first one is mobile application which is used for collecting some data about the user and his/her interests that may be considered as a data entry process. The second module is getting some location information using the GPS of the mobile device. The third module here is matching algorithm, which will do two main functions. The first one is matching the interests of the users and displaying the results based on these interests. The second one is displaying the results and sorting it based on the nearest user.in addition to routing information.

This paper is describing how to add recommendation resources in Blended learning systems. The Blended learning system is a term increasingly used to describe the way e-learning is being combined with traditional classroom methods and independent study to create a new, which called hybrid teaching methodology. It represents a much greater change in basic technique than simply adding computers to classrooms; it represents, in many cases, a fundamental change in the way teachers and students approach the learning experience. This paper describes how an algorithm can be used to make recommendation of resources in the Blended Learning field.

Recommender systems are built to help us to easily find the most proper information on the internet. Unlike the search engines recommender systems bring the information to the user without any manual search effort. This is achieved by using the similarities between users and/or items. There are many methods to build a recommender system and these methods can be applied to many specific domains like shopping, movies and music. Since each application domain has its own specific needs, the method used for recommendations differs. As a specific application domain, news recommender systems aim to give the most relevant news article recommendations to users according to their personal interests and preferences. News recommendation have specific challenges when compared to the other domains. From the technical point of view there are many different methods to build a recommender system. Thus, while general methods are used in news recommendation, researchers also need some new methods to make proper news recommendations. The proposed framework for building automatic recommendations in news is composed of two modules: an off-line module which preprocesses data to build reader and content models, and an online module which uses these models on-the-fly to recognize the reader' needs and goals, and predict a recommendation list. The recommended objects are obtained by using a range of recommendation strategies based mainly on content based filtering and collaborative filtering approaches, each applied separately or in combination.

Student advising is an important and time-consuming effort in academic life. Academic advising has been implemented in order to fill the gap between student and the academic routine, by moving advising, complaining, evaluating, suggesting system from the traditional ways to an automated way. The researcher surveyed the existing literature; as utilized that many institutions have implemented computerized solutions in order to enhance their overall advising experience. In this paper the researcher innovates an automated mechanism for academic advising in the university system. The paper presents an overview of the development and implementation of a new model of e-Academic Advising System as a web-based application. The proposed model attempts to develop a model that the staff and advisor can access to follow-up the student complaints and suggestions. Also, the students who registered can through complain, evaluate & suggest in any subject. Finally, the head of the department can receive a KPIs reports to follow-up his department. Therefore, a need for a system that could detect student’s problems and provide them with suitable feedback is raised. The aim of this paper is to implement a system which facilitates and assists academic advisors in their efforts to providing quality, accurate and consistent advising services to their students; also, to explore the design and implementation of a computerized tool to facilitate this process. This paper discussed the required methodologies used in the development of the Academic Advising System, it has been shown that Academic Advising is a Process more than a Final Product or system, a technical vision for Academic Advising System has been provided. The e-Academic Advising web-based developed and implemented by "Ruby on Rails" as a Web framework which runs via the Ruby programming language and "PostgreSQL" as a Database Engine.

The incredible rising of online networks show that these networks are complex and involving massive data. Giving a very strong interest to set of techniques developed for mining these networks. The clique problem is a wellknown NP-Hard problem in graph mining. One of the fundamental applications for it is the community detection.
It helps to understand and model the network structure which has been a fundamental problem in several fields.
In literature, the exponentially increasing computation time of this problem make the quality of these solutions is
limited and infeasible for massive graphs. Furthermore, most of the proposed approaches are able to detect only
disjoint communities. In this paper, we present a new clique based approach for fast and efficient overlapping
community detection. The work overcomes the short falls of clique percolation method (CPM), one of most
popular and commonly used methods in this area. The shortfalls occur due to brute force algorithm for
enumerating maximal cliques and also the missing out many vertices thatleads to poor node coverage. The proposed work overcome these shortfalls producing NMC method for enumerating maximal cliques then detects overlapping communities using three different community scales based on three different depth levels to assure high nodes coverage and detects the largest communities. The clustering coefficient and cluster density are used to measure the quality. The work also provide experimental results on benchmark real world network to demonstrate the efficiency and compare the new proposed algorithm with CPM method, The proposed algorithm is able to quickly discover the maximal cliques and detects overlapping community with interesting remarks and findings.

This paper aims to estimate the effectiveness of reusing
learning object (LO) by evaluating the most affected aspects
in reusing LO by specialized reviewers. In this study we
propose a metric aims to give very accurate results about the
effectiveness of reusing LO, it can be done by letting a group
of reviewers cooperate in reviewing every LO. Every
reviewer, reviews only his/ her specialized area.
The reviewers of this metric are categorized into three
groups; academic, technical and students while the evaluated
elements are categorized into eight categories; gender,
re-tasking & repurposing, accessibility, appropriateness,
content quality, metadata, motivation and usability.
A comparison to the results of the proposed metric vice the
most famous learning objects' metrics is done through this
study to indicate the reliability and accurateness of the
proposed metric compared with the other metrics.

— Data classification is one of the most important tasks in data mining, which identify to which categories a new observation belongs, on the basis of a training set. Preparing data before doing any data mining is essential step to ensure the quality of mined data. There are different algorithms used to solve classification problems. In this research four algorithms namely support vector machine (SVM), C5.0, K-nearest neighbor (KNN) and Recursive Partitioning and Regression Trees (rpart) are compared before and after applying two feature selection techniques. These techniques are Wrapper and Filter. This comparative study is implemented throughout using R programming language. Direct marketing campaigns dataset of banking institution is used to predict if the client will subscribe a term deposit or not. The dataset is composed of 4521 instances. 3521 instance as training set 78%, 1000 instance as testing set 22%. The results show that C5.0 is superior to other algorithms before implementing FS technique and SVM is superior to others after implementing FS. Keywords— Classification, Feature Selection, Wrapper Technique, Filter Technique, Support Vector Machine (SVM), C5.0, K-Nearest Neighbor (KNN), Recursive Partitioning and Regression Trees (Rpart). I. INTRODUCTION The problem of data classification has numerous applications in a wide variety of mining applications. This is because the problem attempts to learn the relationship between a set of feature variables and a target variable of interest. Excellent overviews on data classification may be found in Classification algorithms typically contain two phases. The first one is training phase in which a model is constructed from the training instances. The second is testing phase in which the model is used to assign a label to an unlabeled test instance[1]. Classification consists of predicting a certain outcome based on a given input. In order to predict the outcome, the algorithm processes a training set containing a set of attributes and the respective outcome, usually called goal or prediction attribute. The algorithm tries to discover relationships between the attributes that would make it possible to predict the outcome. Next the algorithm is given a data set, called prediction set, which contains the same set of attributes, except for the prediction attribute – not yet known. The algorithm analyses the input and produces predicted instances. The prediction accuracy defines how " good " the algorithm is [2]. The four classifiers used in this paper are shown in (figure 1). But many irrelevant, noisy or ambiguous attributes may be present in data to be mined. So they need to be removed because it affects the performance of algorithms. Attribute selection methods are used to avoid over fitting and improve model performance and to provide faster and more cost-effective models [3]. The main purpose of Feature Selection (FS) approach is to select a minimal and relevant feature subset for a given dataset and maintain its original representation. FS not only reduces the dimensionality of data but also enhance the performance of a classifier. So, the task of FS is to search for best possible feature subset depending on the problem to be solved [4]. This paper is organized as follows. Section 2 refers to the four algorithms to deal with the classification problem. Section 3 describes the used FS techniques. Section 4 demonstrates our experimental methodology then section 5 presents the results. Finally section 6 provides conclusion and future work.

—Sentiment analysis is called opinion mining which is the field of study that analyzes people's opinions, sentiments, evaluations, appraisals, attitudes, and emotions towards entities such as products, services, organizations, individuals, issues, events, topics, and their attributes. Starting from the importance of the sentiment analysis generally for individuals and more specifically for gigantic organizations, we started digging in this paper. Graphlab was used to build the sentiment models. Many algorithms were used along with text features selection techniques to predict the positive and negative sentiments like " SVM " , " logistic regression " and " boosted trees ". The mentioned classifiers were applied to a Hotel reviews dataset got from Trip Advisor website to emulate real customer opinions. The results showed that using SVM classifier along with N-grams features selection technique was superior to others.

Social Media and mobile commerce changes the way organizations do business; Social networking has become popular and raised a controversial question on its profitability and future influences. This paper is to provide a general snapshot of social networking and mobile, summarize the benefits and limitations of social commerce and describe mobile social commerce. It also focus on understanding the relationship between m-commerce and e-commerce, discuss the current advantages and disadvantages of ecommerce and m-commerce, identify different m-commerce applications, study of security issues in online marketing and their effect on security and privacy issues. This paper is targeted towards business and IT leaders considering social media and mobile application in some or all of their business applications.

online stores booming up, more and more people today are using online shopping. These networks are complex and involving massive data, giving a very strong interest to a set of techniques developed for mining these graphs enable e-retailers to make better recommendation and comprehend the buying patterns. Community detection is one of the fundamental applications that provide a solution for online stores networks, disciplines where systems are often represented as graphs. It helps to understand and model the network structure, provide a mechanism for executives to assess consumer opinion using this information to improve products, customer service and perception, Also can be useful in various applications such as rating prediction, link and Top-N recommendation, trend analysis and also be the main provider engine for recommender systems. In literature, the exponentially increasing computation time of this problem make the quality of these solutions is limited and impractical. Furthermore, most of the proposed approaches are able to detect only disjoint communities. The paper first focuses on the implementation of new clique based overlapping community detection algorithm, consists of two phases. First to enumerate the maximal cliques. The second to detect the overlapping communities among the discovered maximal cliques using three different community scales based on three different depth levels to discover the largest communities and assures high nodes coverage. While the second part of the work is to rank Top-N nodes through the discovered communities in two ways. The work provides experimental results on Amazon products co-purchasing network, most popular online stores. Clustering coefficient and cluster density are used to measure the quality. The results show that our algorithm can extract meaningful communities in different scales from this network, able to rank the top-N nodes in these different scales and revealing large scale patterns present in interaction habits of customers.

Basically, by increasing Internet of Things (IoT) paradigm involvement in our lives, a lot of threats and attacks regarding IoT security and privacy are realized and raised. If they are left without taking extensive considerations, such security and privacy issues and challenges can certainly threaten the IoT existence. Such security and privacy issues are derived from many reasons. One of these reasons is the unavailability of universal accepted appropriate IoT security techniques, methods, and guidelines. These methods and techniques will greatly guide IoT developers and engineering, draw the success road for developing, and implementing secure IoT systems. So, our contribution focusses on such objective in which, we propose a comprehensive IoT security and privacy framework based on the seven levels of IoT reference architecture introduced by Cisco, in which a set of proper security techniques and guidelines is specified for each level.
Additionally, we identify several critical techniques which can be accomplished for blocking many possible attacks against the IoT seven levels.