Abstract

Apathy is a behavioral and personality change and is generally defined as a loss of motivation. In this study, we will explore the detection of apathy in two phases. An apathy detection phase, and relevant features identification phase.

In the first one, we apply micro-facial expressions detection systems and counter for the purpose of diagnosing clinical manifestation of apathy from a video for Lebanese citizens. The method works by applying Histogram of Oriented Gradients (HOG) as a feature descriptor on video dataset of spontaneous micro facial movements. Micro-facial expressions appear by video recording participants reacting to emotional stimulating COPE cards. Results are compared to Lille Apathy Rating Scale LARS scores. Kappa agreement was calculated to be 95.96% showing the proposed classification method has a high accuracy of estimation.

In a second phase, we aim to identify the demographics and habits that might be affecting the manifestation of apathy using machine leaning algorithms.

A statistical model is built based on the results to identify the characteristics that affect the manifestation of apathy by analyzing the data and making a statistical description. Using a sample of 470 participants, we base our results on the decision tree (CART) combined with logistic regression. Finally, we found that insomnia, genetic background and stress are the most important features that influence the manifestation of apathy, with an accuracy of 96.7%.

Abstract

In a network environment, intrusions pose a serious security problem especially when new intrusion types are discovered which make them difficult to detect. In this work, we use the Classification And Regression Tree (CART) algorithm, a supervised learning technique, on a labeled dataset collected during an attack on the Faculty of Science's web servers, and this in order to classify bad and good connections. As result, the accuracy of the classification reached 99%, maintaining low falsenegative and low false-positive rates.

Abstract

Information flow in logic programming was recently defined using the bottom-up evaluation approach. In this paper, we tackle the same question using the Magic Sets bottom-up evaluation approach. We will show, using this technique that the existence of the information flow is decidable. Finally, we will study the computational complexity of the decidability of the existence of the flow and prove that it is EXPTIME-complete.

Abstract

We extend the concept of bisimilarity relation between datalog goals from positive datalog programs to stratified and restricted Datalog programs with negation. The introduction of negation forced us to reconsider the search space and the semantics in order to guarantee and preserve soundness and completeness results. We address the problem of deciding whether two given goals are bisimilar with respect to a given program. When the given programs are stratified or restricted with negation, this problem is decidable.

Abstract

Information flow in logic programming is well defined using the usual top-down evaluation approach. In this paper, we tackle the same question using a buttom up evaluation approach using two algorithms, Very Naive and Semi Naive algorithms. We will show that, using these techniques, the existence of the information flow is decidable, so that the flow of information can be detected in all cases. Finally, we will study the computational complexities of each decision problem and prove that it's EXPTIME-complete.

Abstract

This paper proposes a formal representation of inference control on information flow theory in logic programming. In order to control the fact that the result returned by a query can convey confidential information, we propose the notion of indistinguishability of flow and elaborate definitions of protection mechanisms, secure mechanisms, precise mechanisms and confidentiality policies based on this notion. We give a secure and precise protection mechanism that prohibits any undesirable inferences and minimizes the number of denials of legitimate actions.

Abstract

This paper presents a new formalization of information flow detection in concurrent logic programming and applies it to the problem of deadlock detection. This work is based on a recent study of the detection of information flow in Datalog programs. Firstly, we define the concept of information flow in concurrent logic programming. Then, we propose a set of definitions of flow based on observation and transition systems while solving goals. Finally, we formalize a mechanism for deadlock detection in concurrent logic programs.

Abstract

Apathy is a behavioral and personality change and is generally defined as a loss of motivation. In this study, we will explore the detection of apathy in two phases. An apathy detection phase, and relevant features identification phase.

In the first one, we apply micro-facial expressions detection systems and counter for the purpose of diagnosing clinical manifestation of apathy from a video for Lebanese citizens. The method works by applying Histogram of Oriented Gradients (HOG) as a feature descriptor on video dataset of spontaneous micro facial movements. Micro-facial expressions appear by video recording participants reacting to emotional stimulating COPE cards. Results are compared to Lille Apathy Rating Scale LARS scores. Kappa agreement was calculated to be 95.96% showing the proposed classification method has a high accuracy of estimation.

In a second phase, we aim to identify the demographics and habits that might be affecting the manifestation of apathy using machine leaning algorithms.

A statistical model is built based on the results to identify the characteristics that affect the manifestation of apathy by analyzing the data and making a statistical description. Using a sample of 470 participants, we base our results on the decision tree (CART) combined with logistic regression. Finally, we found that insomnia, genetic background and stress are the most important features that influence the manifestation of apathy, with an accuracy of 96.7%.

Abstract

This paper proposes a theoretical foundation of what could be an information flow in logic programming. Several information flow definitions (based on success/failure, substitution answers, bisimulation between goals) are stated and compared. Decision procedures are given for each definition and complexity is studied for specific classes of logic programs.

Abstract

This paper deals with the problem of providing assistance to a Bridge player using Artificial Intelligence techniques. In order to achieve this goal, we propose an approach that consists in making assumptions about the honor cards location, then to gradually eliminating the inconsistent assumptions until locating all honor cards. Based on the bids, assumptions about honor cards distribution are made using abduction reasoning. The consistency of these assumptions is managed by an Assumption-based Truth Maintenance System (ATMS). As the game progresses, playing a card generates facts, thereby invalidating some assumptions that cause contradiction and transforming some others into facts corresponding to the location of honor cards.

Abstract

This research provides a method for studying and visualizing different types of relationships in various worlds such as dictionaries and Web pages. As a first step, we identify different types of relationships between objects of the universes previously mentioned. The second step consists in developing a method to quantify the relationship to study. The constituents of the universe (Web pages, items associated with entries in a dictionary ...) form a HSWN graph (hierarchical small world network) whose nodes are entities and edges reflect a direct link (hypertext, definitional ...) between two nodes. This leads to introduce a method for studying the structure of large graphs of hierarchical small world type. Our approach is based on the use of Markovian matrices. A matrix is multiplied k times in order to quantify the relationship between all the nodes of the graph. To illustrate this approach, examples and results on Web and linguistic graphs are given.

Abstract

We introduce the concept of bisimulation between Datalog goals: two Datalog goals are bisimilar with respect to a given Datalog program when their SLD-trees, considered as relational structures, are bisimilar. We address the problem of deciding whether two given goals are bisimilar with respect to given programs. When the given programs are hierarchical or restricted, this problem is decidable in 2EXPTIME.

Abstract

This presentation presents our experience using a novel and innovative approach to teach and assess programming courses, which combines online coding platforms, automated grading systems, interactive visualizations, and offline materials such as registered lessons on YouTube. Our research focused on using these technologies to enhance student engagement and learning outcomes, and we examined the effectiveness of this approach through a comparison of student performance and engagement levels between the new and traditional methods. Our findings demonstrated that students who used the new approach achieved higher grades, reported higher levels of engagement and satisfaction with the courses, and showed an increased demand for more practice, quizzes, and revisions. We also found that students went through a process of refusal, objection, adaptation, and finally acceptance and adoption of the new approach. Overall, our research provides strong evidence that the combination of online and offline materials, along with the use of modern teaching technologies, can significantly enhance the quality of programming courses and contribute to the ongoing revolution in teaching. Our findings have important implications for the field of informatics education and can guide educators in their efforts to adopt innovative and effective teaching methods.

Abstract

This paper presents an innovative strategy for the generation of quizzes and multiple-choice questions (MCQs) within Moodle, leveraging the advanced capabilities of Generative AI through Google's PALM2 models. Designed to substantially streamline educational workflows, our novel Moodle plugin directly addresses the challenges of extensive quiz preparation faced by educators and the necessity for engaging, tailored learning experiences for students. Through rigorous evaluation with both educators and students, our research validates the plugin's efficiency in diminishing administrative tasks and enhancing academic engagement and performance. Key outcomes highlight the plugin's user-friendliness, educational value, and pioneering influence on content generation, leading to a more dynamic and interactive educational environment. This study not only underscores the transformative potential of integrating cutting-edge AI technologies into educational platforms but also lays the groundwork for future investigations into the broad applications of AI in enriching teaching and learning experiences. The broader implications of our findings suggest a vibrant future for digital education, characterized by adaptive, personalized, and enriched learning ecosystems.

Abstract

This paper presents an investigation into machine learning techniques for violence detection in videos and their adaptation to a federated learning context. The study includes experiments with spatio-temporal features extracted from benchmark video datasets, comparison of different methods, and proposal of a modified version of the "Flow-Gated" architecture called "Diff-Gated." Additionally, various machine learning techniques, including super-convergence and transfer learning, are explored, and a method for adapting centralized datasets to a federated learning context is developed. The research achieves better accuracy results compared to state-of-the-art models by training the best violence detection model in a federated learning context.

Abstract

This paper proposes a theoretical foundation of what could be an information flow in logic programming. Several information flow definitions (based on success/failure, substitution answers, bisimulation between resolution trees of goals) are stated and compared. Decision procedures are given for each definition and complexity is studied for specific classes of logic programs.

Abstract

In this study, we investigate the combination of indicators, including performance, behavioral engagement, and emotional engagement, to identify students experiencing difficulties. We analyzed data from two primary sources: digital traces extracted from the Learning Management System (LMS) and images captured by students' webcams. The digital traces provided insights into students' interactions with the educational content, while the images were utilized to analyze their emotional expressions during learning activities. By utilizing real data collected from students at a French engineering school, recorded during the 2022-2023 academic year, we observed a correlation between positive emotional states and improved academic outcomes. These preliminary findings support the notion that emotions play a crucial role in differentiating between high-achieving and low-achieving students.

Abstract

This thesis is developed in order to tackle the issue of information flow in logic programming. The contributions of this thesis can be split into three mains parts:
1* Information flow in logic programming: we propose a theoretical foundation of what could be an information flow in logic programming. Several information flow definitions (based on success/failure, substitution answers, bisimulation between resolution trees of goals) are stated and compared. Decision procedures are given for each definition and complexity is studied for specific classes of logic programs.
2* Bisimulation of logic goals: We introduce the concept of bisimulation between Datalog goals: two Datalog goals are bisimilar with respect to a given Datalog program when their SLD-trees, considered as relational structures, are bisimilar. We address the problem of deciding whether two given goals are bisimilar with respect to given programs. When the given programs are hierarchical or restricted, this problem is decidable in 2EXPTIME.
3* Preventive inference control for deductive databases: We propose a secure and a precise security mechanism for deductive databases based on the notion of information flow in logic programming.