Browsing by Author "Atia, A"

Now showing 1 - 7 of 7

AI-based Flexible Online Laboratory Learning System
(Institute of Electrical and Electronics Engineers Inc., 5/27/2021) Elmesalawy, M.M; Atia, A; Yousef, A.M.F; El-Haleem, A.M.A; Anany, M.G; Elmosilhy, N.A; Salama, A.I; Hamdy, A; Zoghby, H MK; Din, E.S.E
The worldwide outbreak due to COVID-19 pandemic has led to a great interest in e-learning. However, the lack of suitable online laboratory management systems has posed a particular challenge for sectors that need laboratory activities such as engineering, science and technology. In this paper, the requirements and design for a flexible AI-based laboratory learning system (LLS) that can support online laboratory experimentations are presented. The elicitation of the LLS design requirements is decided based on a conducted survey for a set of LLS features. The LLS is designed with the flexibility to support various types of online experimentations such as virtual or remote controlled experiments using either desktop or web applications. The virtualization technique is used to manage the laboratory resources and allow multiple users to access the LLS. Moreover, the proposed LLS introduces the use of AI techniques to provide efficient virtual lab assistant and adaptive assessment process. © 2021 IEEE.
Comparative study for Stylometric analysis techniques for authorship attribution
(5/27/2021) Raafat, M.A; El-Wakil, R.A-F; Atia, A
A text is a meaningful source of information. Capturing the right patterns in written text gives metrics to measure and infer to what extent this text belongs or is relevant to a specific author. This research aims to introduce a new feature that goes more in deep in the language structure. The feature introduced is based on an attempt to differentiate stylistic changes among authors according to the different sentence structure each author uses. The study showed the effect of introducing this new feature to machine learning models to enhance their performance. It was found that the prediction of authors was enhanced by adding sentence structure as an additional feature as the f1_scores increased by 0.3% and when normalizing the data and adding the feature it increased by 5%. © 2021 IEEE.
A Comparative Study of Machine Learning and Deep Learning in Network Anomaly-Based Intrusion Detection Systems
(Institute of Electrical and Electronics Engineers Inc., 2020-12) Abdel-Wahab, M.S; Neil, A.M; Atia, A
This paper presents a comparative study of Machine learning and Deep learning models used in anomaly-based network intrusion detection systems. The paper has presented an overview of the previous work done in the field of ML and DL IDS, then an overview of the used datasets in reviewed literature was presented. Moreover, ML and DL models were tested on the KDD-99 dataset, and performance results were presented, compared, and discussed. Finally, areas of future research of critical importance are proposed by the authors. © 2020 IEEE.
Exploring the Impact of Multimodal Adaptive Learning with Tangible Interaction on Learning Motivation
(Institute of Electrical and Electronics Engineers Inc., 12/15/2020) Chettaoui, N; Atia, A; Bouhlel, M.S
Embodied learning defines a contemporary pedagogical theory focusing on ensuring an interactive learning experience through full-body movement. Within this pedagogy, several studies in Human-Computer Interaction have been conducted, incorporating gestures, and physical interaction in different learning fields. This paper presents the design of a multimodal and adaptive space for embodied learning. The main aim is to give students the possibility to use gestures, body movement, and tangible interaction while interacting with adaptive learning content projected on the wall and the floor. Thus, this study aims to explore how tangible interaction, as a form of implementing embodied learning, can impact the motivation of students to learn compared to tablet-based learning. Eighteen primary school students aged nine and ten years old participated in the study. The average percentages of answers on the Questionnaire on Current Motivation (QCM) pointed out a higher motivation among students learning via tangible objects. Results revealed a positive score for the Interest of learning abstract concepts using a tangible approach with a mean score of 4.78, compared to 3.77 while learning via a tablet. Furthermore, Success and Challenge measures, with a mean score of 4.67 and 4.56 indicate that physical interaction via tangible objects leads to significantly higher motivation outcomes. These findings suggest that learning might benefit more from a multimodal and tangible physical interaction approach than the traditional tablet-based learning process. © 2020 IEEE
IoT System Based on parameter optimization of Deep Learning using Genetic Algorithm
(Intelligent Network and Systems Society, 2021-03) Slim, S.O; Elfattah, M.M.A; Atia, A; Mostafa, M.-S.M
View references (73) Nowadays, more and more human activity recognition (HAR) tasks are being solved with deep learning techniques because it’s high recognition rate. The architectural design of deep learning is a challenge because it has multiple parameters which effect on the result. In this work, we propose a novel method to enhance deep learning architecture by using genetic algorithm and adding new statistical features. Genetic algorithm is utilized as an enhancing method to get the optimal value parameters of deep learning. Also new statistical features are appended to the features that are extracted automatically from CNN technique. Because the spread of the internet and its significance in our life, we developed Internet of Things (IoT) system. Therefore, we evaluated the performance of the proposed method in its system and found satisfactory results. Moreover, the proposed method was trained on two benchmark datasets (WISDM and UCI) and tested on the dataset, which was collected from IoT system. The results showed that the proposed model improved the accuracy up to 93.8% and 86.1% for user-dependent and independent. © 2020
Multi-Sensor Fusion for Online Detection and Classification of Table Tennis Strokes
(Intelligent Network and Systems Society, 2021-03) Hegazy, H; Abdelsalam, M; Hussien, M; Elmosalamy, S; Hassan, Y; Nabil, A; Atia, A
Sports training generally focuses on speed of response and variety of strategies aimed at encouraging sustainable physical activity and improving learning skills for players, thus, enhance their performance and skills through matches. In this paper, we are presenting a methodology for multi-sensor fusion method of IR depth camera and smart band sensors using curve fitting to recognize and group the characteristics of strokes performed by players and adapt the training accordingly. The main aim of the methodology is to classify various techniques played in table tennis and enhance the strokes based on various body joints. Moreover, the main contribution of this paper is to experiment different sensors to get the most optimal tools in classifying players strokes. Also, to test various classification algorithms to get the optimal and best result possible. Overall, based on the experiments we have concluded that sensor fusion between internal sensors and IR depth camera has increased the classification results and robustness of the solution. The system’s results indicate an average accuracy of 95% - 100%. © 2020
Predicting Student Performance in an Embodied Learning Environment
(Institute of Electrical and Electronics Engineers Inc., 5/27/2021) Chettaoui, N; Atia, A; Bouhlel, M.S
Implementing machine learning techniques is receiving considerable attention in the educational technology research field. Different systems and techniques were proposed to predict student performance and gain insights regarding their learning needs. However, to the best of our knowledge, no previous research studies explored predicting student performance based on their interaction with different modalities in a real classroom context In this paper, we have proposed an adaptive learning engine to recommend embodied activities and interaction modalities according to students learning profile, prior knowledge, and academic performance. We have firstly conducted a comparative study by analyzing two educational datasets. Among five classification algorithms, Random Forest outperformed with 84% as accuracy. Thus, in a second phase, we have explored predicting students' performance with our dataset generated from embodied learning environment. The accuracy of the proposed model using student profile, academic, and behavioral features reached an accuracy of 62%. These results highlighted the potential of implementing machine learning techniques to classify and recommend embodied interaction