This study examined the integration of artificial intelligence-powered speech recognition technology within early reading assessments in Morocco’s Teaching at the Right Level (TaRL) program. The purpose was to evaluate the effectiveness of an automated speech recognition tool compared to traditional paper- based assessments in improving reading skills among 100 Moroccan first to third-graders. The mixed- method approach combined pre-post standardized reading tests with qualitative feedback. Results showed students receiving the AI-enabled speech recognition assessments demonstrated significant gains in reading achievement compared to peers assessed via traditional methods. Qualitative findings revealed benefits of instant feedback and enhanced engagement provided by the speech recognition tool. This study contributes timely empirical evidence on adopting learning technologies, specifically AI-driven automated speech assessment instruments, to enhance foundational literacy development within under-resourced education systems implementing student-centered pedagogical techniques like TaRL. It provides valuable insights and guidance for integrating innovative speech analysis tools within localized teaching and learning frameworks to strengthen early reading instruction and monitoring.

In this era of Artificial Intelligence (AI) growth, characterized by advances in the Large Language Models (LLMs) used by ChatGPT and Bard, this study examines the effects of gamification and Automatic Question Generation (AQG) on student engagement and learning outcomes in the context of a Massive Open Online Course (MOOC). AQG, implemented via a Moodle plugin, transforms conventional assessments into an interactive, gamified experience, leveraging the “test effect” to improve learning outcomes. Research with 100 fifth-graders in a primary and secondary school shows that gamified assessments significantly boost student motivation and learning outcomes compared with traditional methods. The custom Moodle plugin facilitates the AQG process, generating contextually relevant and grammatically correct Multiple-Choice Questions (MCQs) from course content. The result is a dynamic, personalized assessment experience aimed at optimizing student retention. This paper concludes by discussing the implications of the study for educators and highlighting potential directions for future research.

The aim of this research is to provide a novel educational model with the goal of reducing the expenses associated with manual question production and meeting the demand for a continual supply of new questions on MOOC platforms such as Moodle or Open EDX. We considered integrating machine-learning methods with natural language processing in order to increase the number and validity of assessing questions. To accomplish this, we developed a system that generates multilingual questions automatically.

Various kinds of evaluation were conducted with two factors in mind: evaluating MOOC learners' competency and the similarity of the generated questions to those created by humans. The first evaluation is based on subjective judgment by three MOOC creators, while the second is based on replies from MOOC participants on machine-generated and human-created questions. Both evaluations revealed that the machine-generated questions performed on par with the human-created questions in terms of evaluating skills and similarity. Moreover, the results demonstrate that most of the produced questions (up to 82 percent) enhance e-assessment when the new suggested technology is used.

Today, e-learning has emerged as a key issue, with applications in every business, including supported training, lifetime learning modes, and support tools for all types of educational institutions. As a result of the COVID-19’s impact, millions of applicants take online tests via MOOC platforms, and while scoring brief responses is crucial to the process, it becomes difficult to grade them using human graders. Unfortunately, multiple-choice quizzes are inadequate for measuring higher-level learning that extends beyond the lower levels of Bloom’s taxonomy. As a result, developing and implementing a generic automated system for grading these responses is critical. Utilizing computers to evaluate and grade students’ assignments simplifies and cuts costs. We investigate an automated answer scoring technique that combines machine learning techniques with automatic natural language processing to produce a fast, scalable, and accurate result.

In today’s generation of MOOCs, videos are fundamental to the student learning experience. Due to the prominence of video content in MOOCs, production staff and instructional designers invest significant time and resources in creating these videos. With instructional videos, they want to increase student involvement. Without formative assessment, however, actual engagement is hard to quantify. Nonetheless, a large number of students necessitates a larger pool of questions; to address this issue, we considered mixing machine-learning methods with automatic natural language processing in order to expand the number of questions evaluated and ensure their validity. To accomplish this, we implemented a methodology that generates questions automatically from video transcripts. Following each course comes an evaluation issue, which is typically a multiple-choice question designed to measure a student's comprehension of the video's material. machine-generated questions performed comparably to human-generated questions when it came to judging skill and resemblance. Additionally, the findings indicate that the majority of the questions generated improve e-assessment when the new technology is applied.

This study introduces a generative AI-based Flashcard Bot, utilizing natural language processing (NLP) for flashcard creation within the Massive Open Online Courses (MOOCs) context. By leveraging the pre-trained T5 model, the system transforms the extracted text content of a business intelligence MOOC from Moodle into flashcards. These AI-generated flashcards, aimed at enhancing learners’ memory through testing effect and spaced repetition, are evaluated based on quality, relevance, accuracy, and fluency. Comparisons are made with existing flashcard tools, and student perceptions are assessed through a survey. The results illustrate comparable quality between AI-generated and human-created flashcards, signifying the potential of generative AI and NLP in transforming educational technology and augmenting online learning experiences.

The university must adapt to the challenges which have characterized the 21st century. According to UNESCO, during the period of the COVID-19 pandemic the number of learners has been affected by school closures; it has exceeded 1.5 billion in 195 countries. The Sultan Moulay Slimane University in Morocco has employed a strategy through distance education which makes the online courses massive and open to all, but there are many obstacles such as the problem of choice due to the diversity of platforms and multilingual e-assessment tools. To solve this serious problem, we have thought of establishing criteria for choosing a suitable learning management system and integrating new automatic natural language processing, using artificial intelligence techniques to make MOOCs more attractive. To do this, we created a plugin of automatic multilingual questions generation which transforms the course into an addictive game with points and badges. To solve this serious problem, we have thought of establishing criteria for choosing a suitable learning management system and integrating new automatic natural language processing, using artificial intelligence techniques to make MOOCs more attractive. To do this, we created a plugin of automatic multilingual questions generation which transforms the course into an addictive game with points and badges.

The COVID-19 pandemic has necessitated a rapid shift towards distance learning, presenting unique challenges and opportunities for educational institutions worldwide. Sultan Moulay Slimane University has responded to these challenges by pioneering the integration of artificial intelligence (AI) to enhance hybrid learning models. This study explores the development and implementation of AI-driven methodologies for automatic question generation from video transcripts, with the aim of improving student engagement and assessment in distance learning environments. Utilizing state-of-the-art machine learning algorithms and natural language processing (NLP) techniques, we demonstrate the potential of AI to facilitate personalized, efficient, and effective e-learning experiences. Our findings reveal that AI-generated questions not only match but, in some instances, surpass the quality of human-generated counterparts in terms of accuracy, relevance, and the ability to diversify assessment strategies. This investigation underscores the transformative impact of AI in educational settings, offering insights into its role in fostering adaptive and inclusive learning environments in the post-pandemic era. Furthermore, the study addresses critical research questions concerning the optimization of NLP for educational content, the development of robust models for educational question creation, and the utilization of AI-generated assessments to reduce instructor workload while enhancing personalization and engagement in hybrid learning contexts. By providing data-driven insights and frameworks, this research aims to guide the adoption of AI for next-generation online pedagogies, assessments, and learning experiences, highlighting the potential of human-AI collaboration to transform assessment in online education. 

The COVID-19 pandemic has necessitated a rapid shift towards distance learning, presenting unique challenges and opportunities for educational institutions worldwide. Sultan Moulay Slimane University has responded to these challenges by pioneering the integration of artificial intelligence (AI) to enhance hybrid learning models. This study explores the development and implementation of AI-driven methodologies for automatic question generation from video transcripts, with the aim of improving student engagement and assessment in distance learning environments. Utilizing state-of-the-art machine learning algorithms and natural language processing (NLP) techniques, we demonstrate the potential of AI to facilitate personalized, efficient, and effective e-learning experiences. Our findings reveal that AI-generated questions not only match but, in some instances, surpass the quality of human-generated counterparts in terms of accuracy, relevance, and the ability to diversify assessment strategies. This investigation underscores the transformative impact of AI in educational settings, offering insights into its role in fostering adaptive and inclusive learning environments in the post-pandemic era. Furthermore, the study addresses critical research questions concerning the optimization of NLP for educational content, the development of robust models for educational question creation, and the utilization of AI-generated assessments to reduce instructor workload while enhancing personalization and engagement in hybrid learning contexts. By providing data-driven insights and frameworks, this research aims to guide the adoption of AI for next-generation online pedagogies, assessments, and learning experiences, highlighting the potential of human-AI collaboration to transform assessment in online education. 

The evaluation of learners in MOOCs (Massive Open Online Courses) poses considerable challenges linked to scaling up and the heterogeneity of audiences. This thesis aims to optimize the evaluation in this context using artificial intelligence algorithms. More specifically, Deep Learning and Natural Language Processing (NLP) techniques are explored. Automatic generation of multiple choice questions based on pre-trained language mod- els helps produce quality assessment quizzes. Text feature extraction and the use of machine learning models make automatic scoring of essays possible with high accuracy. Gamification, through the integration of fun elements and rewards, increases student motivation and engagement. Finally, automatic generation of flashcards using natural language processing techniques proves to be as effective as manual creation in improving retention. Teaching at the Right Level (TaRL) driven by Automatic Speech Recognition (ASR) can be extended to MOOCs. This approach offers significant benefits for language learning and people with disabilities. The results obtained demonstrate the significant potential of these approaches based on artificial intelligence. Automatically generated questions are comparable to those generated by humans. Automatic essay scoring achieves an average accuracy of 0.93 according to the Kappa score. Gamification clearly increases student motivation. Auto- matically generated flashcards are similar in quality and utility to those made manually. The extension of TaRL driven by ASR to MOOCs improves reading performance and the personalization of teaching. This thesis provides evidence that AI can significantly improve assessment in MOOCs, making it more effective, objective and engaging for learners. The techniques presented have the potential to transform assessment in online education.

This thesis describes the development of a software application called GALY v2.0 for automated management of high schools. It was created by computer science student Younes-Aziz Bachiri as an internship project for the Regional Academy of Education and Training of Marrakesh-Tensift-Al Haouz. The goal of the project was to computerize the information system of high schools to simplify workflow related to continuous assessment grades. The application includes functionality for school administrators and teachers to manage student records, enter grades, assign students to classes, generate class lists, print transcripts, and produce statistics. The project involved requirements analysis, UML modeling, database design with MERISE methodology, and development in Java using NetBeans. JasperReports was used for reporting features. Testing, user documentation, and training were also completed. The first version was officially released in May 2011. Overall, the project provided good experience in software analysis, design, development, and delivery. It gave the students real insight into the role of a software analyst and programmer, including managing constraints, priorities, and deadlines. They were satisfied to deliver a reliable application that can streamline operations for the schools and Regional Academy.

This thesis describes the development of a distributed application for vehicle localization, monitoring, and alerting. The project was done for the Faculty of Science and Technology in Beni Mellal, Morocco for their fleet of vehicles. The overall mission was divided into two parts - conducting an in-depth study of the project's business aspects and technologies, and then building the actual application. For the second part, the work involved designing, developing, and deploying the client/server application. It uses GPS tracking devices installed in the vehicles to collect location data and send it via GPRS to the server. The web-based application allows real-time vehicle tracking on maps as well as viewing past trips and generating reports. It also has functionality for user account management, vehicle/driver management, geo-fencing, and notifications. Various technologies were employed including Java, Spring Boot, MySQL, Hibernate, REST, and AngularJS. Overall, the project provided good experience in analysis, design, development, and delivery of a distributed GPS fleet tracking solution using current technologies. Future work could involve adding features like SMS alerts or enhancing reporting capabilities.