Elhoucine OUASSAM

Elhoucine OUASSAM


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 February 2024 in Baghdad Science Journal
Verified
Journal Article
Open access
PDF file
EOElhoucine OUASSAM
Elhoucine OUASSAM
YDYassine Dabachine
Yassine Dabachine
NHNabil Hmina
Nabil Hmina
BBBelaid Bouikhalene
Belaid Bouikhalene
يمكن أن يؤدي استخدام التعلم الآلي في الوقت الفعلي لتحسين إجراءات مراقبة جوازات السفر في المطار إلى تحسين كفاءة العملية وأمانها بشكل كبير. لأتمتة هذه الإجراءات وتحسينها ، يمكن تنفيذ خوارزميات الذكاء الاصطناعي مثل التعرف على الأحرف والتعرف على الوجه والخوارزميات التنبؤية والمعالجة التلقائية للبيانات. تتمثل الطريقة المقترحة في استخدام نموذج R-CNN للكشف عن الأشياء لاكتشاف أجسام جواز السفر في الصور في الوقت الفعلي التي تم جمعها بواسطة كاميرات مراقبة الجوازات. تصف هذه المقالة العملية خطوة بخطوة للنهج المقترح ، والتي تشمل المعالجة المسبقة والتدريب واختبار نموذج R-CNN ودمجه في نظام مراقبة جوازات السفر وتقييم دقته وموثوقيته وسرعة الإدارة الفعالة لـتدفقات الركاب في المطارات الدولية. وقد أظهر تطبيق هذه الطريقة أداءً فائقًا مقارنة بالطرق السابقة من حيث تقليل الأخطاء والتأخيرات والتكاليف المرتبطة بها.

18 April 2024
Verified
Conference Paper
Open access
HAHamzaoui
Hamzaoui
BABachiri
Bachiri
EOElhoucine OUASSAM
Elhoucine OUASSAM
MOMouncif
Mouncif
BOBouikhalene
Bouikhalene
The COVID-19 pandemic has necessitated a rapid shift to online and hybrid learning models, presenting unique challenges and opportunities in the educational sector. This study, conducted at Sultan Moulay Slimane University (USMS) in Morocco, explores the integration of Artificial Intelligence (AI) to enhance distance learning through the automatic generation of assessment questions from video content. By leveraging advanced natural language processing (NLP) and deep learning techniques, the research aims to address the limitations of passive video watching and promote active learning and self-regulated learning skills. The methodology encompasses the use of speech recognition to transcribe video content, summarization techniques to distill core concepts, and a Text2Text generation model for creating relevant assessment questions. The study's findings demonstrate that the AI-driven question generation system can match the quality, diversity, and relevance of human-authored questions, thereby reducing instructor workload and fostering personalization and engagement in hybrid learning contexts. This research contributes to the body of knowledge on applying AI in education, offering insights and frameworks for adopting AI-driven methodologies to improve online pedagogies, assessments, and learning experiences in the post-pandemic era.

1 September 2023 in Indonesian Journal of Electrical Engineering and Computer Science
Verified
Journal Article
Open access
PDF file
EOElhoucine OUASSAM
Elhoucine OUASSAM
NHNabil Hmina
Nabil Hmina
BBBelaid Bouikhalene
Belaid Bouikhalene
Assessment is a key element in today’s School system, whether face-to-face or distance learning, as it helps students understand their learning and get feedback on their progress. In addition, distance learning assessment is becoming increasingly popular as it is convenient for students with busy schedules who cannot attend face-to-face assessments. In this paper, we focus on the use of intelligent agents on the Moodle platform to improve the assessment process of distance learning. We present three contributions that aim to improve the developed models: firstly, the digitisation of assessment to collect, store and analyse data; secondly, the adoption of a multi-agent skills assessment environment to automate some assessment tasks; thirdly, the adoption of the leadership and management development (LMD) programme to improve the continuous training of learners by offering greater flexibility, adaptability and relevance to their needs.

30 August 2025
Verified
Book Chapter
Closed access
PDF file
EOElhoucine OUASSAM
Elhoucine OUASSAM
HRHamzaoui Radoine
Hamzaoui Radoine
YBYounes-Aziz Bachiri
Younes-Aziz Bachiri
BBBouikhalene Belaid
Bouikhalene Belaid

23 February 2023 in Research Square
Verified
Preprint
Open access
PDF file
EOElhoucine OUASSAM
Elhoucine OUASSAM
YDYassine Dabachine
Yassine Dabachine
NHNabil Hmina
Nabil Hmina
BBBelaid Bouikhalene
Belaid Bouikhalene
This work presents a game-theoretic approach to address the problem of coordination and scheduling of aircraft at intersections, with the goal of avoiding conflicts and potential collisions. The proposed algorithm enables simple agents to work together in a way that leads to cooperative behaviors, resulting in equilibria that improve the overall efficiency of the system. The researchers tested and compared the game-theoretic approach with a centralized approach, specifically First-Come-First-Serve (FCFS), using data from Mohammed 5 Casablanca airport. The initial results suggest that the game-theoretic model is promising, despite its higher complexity. The approach has the potential to improve the overall coordination and scheduling of aircraft, leading to a more efficient and safe system. The proposed game-theoretic approach is designed to improve the coordination and scheduling of aircraft at intersections, ultimately leading to a safer and more efficient system. The approach is shown to be promising in initial testing, offering a potentially superior alternative to centralized approaches like FCFS. This research highlights the potential benefits of game-theoretic models in addressing complex coordination problems in multi-agent systems.

5 October 2023
Verified
Conference Paper
Closed access
EOElhoucine OUASSAM
Elhoucine OUASSAM
NHNabil Hmina
Nabil Hmina
BBBelaid Bouikhalene
Belaid Bouikhalene
Hanaa HachimiHanaa Hachimi
Hanaa Hachimi
In this paper, we presented a review of the literature on the application of distributed and agentified heuristic methods to complex systems in AI. We begin by introducing the key concepts of distribution and agentification, and the benefits and challenges associated with their use in complex systems. Then, we review some heuristic approaches that have been developed to solve complex problems in areas such as optimization, planning, pattern recognition, learning, etc. In particular, we focus on heuristic methods that can be distributed and/or agentic, highlighting their main features and the results obtained in a concrete application. Finally, we discuss the future prospects and challenges of distributing and agenticating heuristic methods in AI.