Advances in Information, Communication and Cybersecurity, 2022
Machine learning techniques are compressively used in various fields like business, health, educa... more Machine learning techniques are compressively used in various fields like business, health, education, etc. The growing necessity in the education segment has concreted the approach for plenty of research projects that strongly emphasize student academic achievement and behavior analysis. This research study is about an analysis performed to identify the significant tuition waiver significance of student academic background, accomplishment, and family background. Supervised machine learning approaches gradually, Decision Tree Regression (DTR) and Random Forest Regression (RFR) are utilized in predicting tuition waivers. In addition, Cross-Validation (CV) is used for data overfitting. For better accuracy, DTR and RFR are implemented twice before and after cross-validation. There is enough data to train the models thoroughly and get quite good accuracy and performance. Results appear that the accuracy of DTR and RFR are 74.49% and 77.82%, respectively for before applied CV; and, 71.41%, and 76.90% respectively, for after applied CV.
The student’s performance plays an important role in producing the best quality graduate who will... more The student’s performance plays an important role in producing the best quality graduate who will responsible for the country’s economic growth and social development. The labor market also concerns with student’s performance because the fresh graduate students are considered as an employee depends on their academic performance. So, identification of the reason behind student’s performance variation provides valuable information for planning education and policies. Many researchers try to find out the reason with different types of data mining approaches in different countries. However, none of them worked with Bangladeshi students. This paper proposed a model for identifying the key factors of variation Bangladeshi students’ academic performance and predicts their results. This paper proposes a model which able to identify the students who need special attention. Different types of feature selection methods were used such as Co-relation, Chi-Square and Euclidean distance to select valuable features and feature selections result through decision tree, Naive Bayes, K-nearest neighbor and Artificial Neural Network classifiers algorithm were compared. The performance analysis is done by using student SGPA and review on given facilities from a university. From the performance analysis result it is found that, decreasing number of classes in dataset, the Artificial Neural Network (ANN) (93.70%) performs better than Decision Tree (DT) (92.18%), K-Nearest Neighbors (KNN) (77.74%) and Naive Bayes (NB) (68.33%). However, an increasing number of classes in dataset the DT perform better than ANN, KNN, NB.
2019 IEEE Student Conference on Research and Development (SCOReD), 2019
Kidnapping and harassment is not only a global issue but also a historic issue in Bangladesh. In ... more Kidnapping and harassment is not only a global issue but also a historic issue in Bangladesh. In between the years of 2010 and 2018, the total number of kidnapping events were found, 6708 (Avg. 745.33 events per year) in Bangladesh. The government is trying to capture and punish the kidnappers. But there are a few ways by which the victim can also notify the responsible persons about the unwanted incidents in real-time. This research proposes to design and develop an anti-kidnapping and anti-harassment mobile application that is consists of two modules, the Avoidance and the Detection modules. The Avoidance module notifies the user about some unsafe locations. The Detection module is further divided into four submodules, the Notification module, Sound module, Sensor module, and Spy module. The Notification module is activated by pressing the SOS key and the Sound module is activated by voice command. Both of these modules can send the current location of the user by sending SMS to some selected contact numbers. In the Voice module, the audio voice can be recorded for 15 seconds and saved in phone storage or cloud. The Sensor module uses accelerometer sensor and compass sensor. During unwanted circumstances, the accelerometer sensor is activated by shaking the phone three times and the compass sensor is activated by the movement of the user. The Notification module is tested by sending notifications in different areas in Bangladesh. It is found that the average response time of the notification module is 0.74 milliseconds. The complete application is tested by following the System Usability Scale (SUS) method. It is found that the SUS score of the system is 75.56 %, which indicates that the system is good enough to use.
Advances in Intelligent Systems and Computing, 2021
–Nowadays people feel more comfortable to use Android phone for their regular day-to-day activiti... more –Nowadays people feel more comfortable to use Android phone for their regular day-to-day activities. Besides, it creates an easy access to achieve any goal in a tight schedule. Along that, affordable smart-phone technologies are introducing newer communication possibilities that were not imaginable a few years back. Considering the impact of this Android technology on the students, we have designed and constructed an Android application named ‘IUB SECS’. The objective was to build an effective application for students and faculty members of Independent University, Bangladesh to accomplish specific educational goals and duties by using an Android application. By using this application, students can find course information, lecture notes, faculty information, research information and updated results from outside the campus. On the other hand, faculty members can use it for online attendance, updating course materials, lecture notes and information related to the course and university. The application will provide users the opportunity to create a database, where profile of faculty members and students can be stored and through it students can directly contact with the faculty members. The usability and effectiveness of this application have been evaluated based on five different users. The practical result shows that the outcome of this application is very promising. Keywords––Android application, university, faculty, customize, lecture, course, attendance, grades
... UTILIZATION OF SOLAR EARTH-WATER STILLS FOR DESALINATION OF GROUNDWATER AN MINASIAN, AA AL-KA... more ... UTILIZATION OF SOLAR EARTH-WATER STILLS FOR DESALINATION OF GROUNDWATER AN MINASIAN, AA AL-KARAGHOULI, M. HASAN, and A. SHAKIR Solar ... Each Ahmadzadeh [ 3 ] reported the effects of initial soil hole had an area of ( 1.0 0.75 ) m2 and was 1.2 m ...
Among different color models HSV, HLS, YIQ, YCbCr, YUV, etc. have been most popular for skin dete... more Among different color models HSV, HLS, YIQ, YCbCr, YUV, etc. have been most popular for skin detection. Most of the research done in the field of skin detection has been trained and tested on human images of African, Mongolian and Anglo-Saxon ethnic origins, skin colors of Indian sub-continentals have not been focused separately. Combinatorial algorithms, without affecting asymptotic complexity can
Road surface monitoring is mostly done manually in cities which is an intensive process of time c... more Road surface monitoring is mostly done manually in cities which is an intensive process of time consuming and labor work. The intention of this paper is to research on road damage detection and classification from road surface images using object detection method. This paper applied multiple convolutional neural network (CNN) algorithm to classify road damage and discovered which algorithm performs better in road damage detection and classification. The damages are classified in three categories pothole, crack and revealing. For this research data was collected from street of Dhaka city using smartphone camera and prepossessed the data like image resize, white balance, contrast transformation, labeling. This study applies R-CNN and faster R-CNN for object detection of road damages and apply Support Vector Machine (SVM) for classification and gets a better result from previous studies. Then losses are calculated using different loss functions. The results demonstrate the highest 98.88% accuracy and the lowest loss is 0.01.
Advances in Information, Communication and Cybersecurity, 2022
Machine learning techniques are compressively used in various fields like business, health, educa... more Machine learning techniques are compressively used in various fields like business, health, education, etc. The growing necessity in the education segment has concreted the approach for plenty of research projects that strongly emphasize student academic achievement and behavior analysis. This research study is about an analysis performed to identify the significant tuition waiver significance of student academic background, accomplishment, and family background. Supervised machine learning approaches gradually, Decision Tree Regression (DTR) and Random Forest Regression (RFR) are utilized in predicting tuition waivers. In addition, Cross-Validation (CV) is used for data overfitting. For better accuracy, DTR and RFR are implemented twice before and after cross-validation. There is enough data to train the models thoroughly and get quite good accuracy and performance. Results appear that the accuracy of DTR and RFR are 74.49% and 77.82%, respectively for before applied CV; and, 71.41%, and 76.90% respectively, for after applied CV.
The student’s performance plays an important role in producing the best quality graduate who will... more The student’s performance plays an important role in producing the best quality graduate who will responsible for the country’s economic growth and social development. The labor market also concerns with student’s performance because the fresh graduate students are considered as an employee depends on their academic performance. So, identification of the reason behind student’s performance variation provides valuable information for planning education and policies. Many researchers try to find out the reason with different types of data mining approaches in different countries. However, none of them worked with Bangladeshi students. This paper proposed a model for identifying the key factors of variation Bangladeshi students’ academic performance and predicts their results. This paper proposes a model which able to identify the students who need special attention. Different types of feature selection methods were used such as Co-relation, Chi-Square and Euclidean distance to select valuable features and feature selections result through decision tree, Naive Bayes, K-nearest neighbor and Artificial Neural Network classifiers algorithm were compared. The performance analysis is done by using student SGPA and review on given facilities from a university. From the performance analysis result it is found that, decreasing number of classes in dataset, the Artificial Neural Network (ANN) (93.70%) performs better than Decision Tree (DT) (92.18%), K-Nearest Neighbors (KNN) (77.74%) and Naive Bayes (NB) (68.33%). However, an increasing number of classes in dataset the DT perform better than ANN, KNN, NB.
2019 IEEE Student Conference on Research and Development (SCOReD), 2019
Kidnapping and harassment is not only a global issue but also a historic issue in Bangladesh. In ... more Kidnapping and harassment is not only a global issue but also a historic issue in Bangladesh. In between the years of 2010 and 2018, the total number of kidnapping events were found, 6708 (Avg. 745.33 events per year) in Bangladesh. The government is trying to capture and punish the kidnappers. But there are a few ways by which the victim can also notify the responsible persons about the unwanted incidents in real-time. This research proposes to design and develop an anti-kidnapping and anti-harassment mobile application that is consists of two modules, the Avoidance and the Detection modules. The Avoidance module notifies the user about some unsafe locations. The Detection module is further divided into four submodules, the Notification module, Sound module, Sensor module, and Spy module. The Notification module is activated by pressing the SOS key and the Sound module is activated by voice command. Both of these modules can send the current location of the user by sending SMS to some selected contact numbers. In the Voice module, the audio voice can be recorded for 15 seconds and saved in phone storage or cloud. The Sensor module uses accelerometer sensor and compass sensor. During unwanted circumstances, the accelerometer sensor is activated by shaking the phone three times and the compass sensor is activated by the movement of the user. The Notification module is tested by sending notifications in different areas in Bangladesh. It is found that the average response time of the notification module is 0.74 milliseconds. The complete application is tested by following the System Usability Scale (SUS) method. It is found that the SUS score of the system is 75.56 %, which indicates that the system is good enough to use.
Advances in Intelligent Systems and Computing, 2021
–Nowadays people feel more comfortable to use Android phone for their regular day-to-day activiti... more –Nowadays people feel more comfortable to use Android phone for their regular day-to-day activities. Besides, it creates an easy access to achieve any goal in a tight schedule. Along that, affordable smart-phone technologies are introducing newer communication possibilities that were not imaginable a few years back. Considering the impact of this Android technology on the students, we have designed and constructed an Android application named ‘IUB SECS’. The objective was to build an effective application for students and faculty members of Independent University, Bangladesh to accomplish specific educational goals and duties by using an Android application. By using this application, students can find course information, lecture notes, faculty information, research information and updated results from outside the campus. On the other hand, faculty members can use it for online attendance, updating course materials, lecture notes and information related to the course and university. The application will provide users the opportunity to create a database, where profile of faculty members and students can be stored and through it students can directly contact with the faculty members. The usability and effectiveness of this application have been evaluated based on five different users. The practical result shows that the outcome of this application is very promising. Keywords––Android application, university, faculty, customize, lecture, course, attendance, grades
... UTILIZATION OF SOLAR EARTH-WATER STILLS FOR DESALINATION OF GROUNDWATER AN MINASIAN, AA AL-KA... more ... UTILIZATION OF SOLAR EARTH-WATER STILLS FOR DESALINATION OF GROUNDWATER AN MINASIAN, AA AL-KARAGHOULI, M. HASAN, and A. SHAKIR Solar ... Each Ahmadzadeh [ 3 ] reported the effects of initial soil hole had an area of ( 1.0 0.75 ) m2 and was 1.2 m ...
Among different color models HSV, HLS, YIQ, YCbCr, YUV, etc. have been most popular for skin dete... more Among different color models HSV, HLS, YIQ, YCbCr, YUV, etc. have been most popular for skin detection. Most of the research done in the field of skin detection has been trained and tested on human images of African, Mongolian and Anglo-Saxon ethnic origins, skin colors of Indian sub-continentals have not been focused separately. Combinatorial algorithms, without affecting asymptotic complexity can
Road surface monitoring is mostly done manually in cities which is an intensive process of time c... more Road surface monitoring is mostly done manually in cities which is an intensive process of time consuming and labor work. The intention of this paper is to research on road damage detection and classification from road surface images using object detection method. This paper applied multiple convolutional neural network (CNN) algorithm to classify road damage and discovered which algorithm performs better in road damage detection and classification. The damages are classified in three categories pothole, crack and revealing. For this research data was collected from street of Dhaka city using smartphone camera and prepossessed the data like image resize, white balance, contrast transformation, labeling. This study applies R-CNN and faster R-CNN for object detection of road damages and apply Support Vector Machine (SVM) for classification and gets a better result from previous studies. Then losses are calculated using different loss functions. The results demonstrate the highest 98.88% accuracy and the lowest loss is 0.01.
Uploads
Papers