French 
English 
Current Masters
View detailed information
Enhancing Retrieval-Augmented Generation (RAG)-Based Personalized Mindfulness Cognitive Therapy with a Reranking Approach
Supervisor: Amine Boufaied
My research focuses on the development of an intelligent chatbot for mental health support, particularly mindfulness. I leverage Retrieval-Augmented Generation (RAG) systems combined with Large Language Models (LLMs), with an enhanced retrieval process through an emotional reranking strategy, in order to provide more relevant and empathetic responses
Using large languge Models for gene Annotation
Supervisor: Mehdi Azaouzi
My thesis focuses on the use of Large Language Models (LLMs) for functional gene annotation. The aim is to develop an approach capable of automatically predicting the biological functions of genes from their sequences, leveraging biomedical corpora. This method seeks to improve the accuracy, speed, and scope of genomic annotation, thereby contributing to a better understanding of biological processes and the advancement of precision medicine.
An LLMs-based Chatbot for Mental Health
Supervisor: Lotfi Ben Romdhane
My master’s thesis focuses on the use of Large Language Models (LLMs) in the field of mental health. This work explores the integration of emotional information into the attention mechanism of LLMs in order to enhance their understanding of human language in a psychological context. The study is based on a technical adaptation of Transformer-based architectures to sensitive mental health data.
Genome Variant Classifation Using Large Language Models
Supervisor: Youssef Gamha
This project aims to develop a model based on Large Language Models (LLMs) for the classification of genomic variants. The objective is to improve the accuracy of genetic diagnosis through the automatic interpretation of mutations. This work falls within the fields of personalized medicine and bioinformatics.
Recherche d'information pertinentes basé sur les réseaux sociaux
Supervisor: Youssef Gamha
Mon sujet de recherche porte sur la conception d’un système intelligent de Recherche d’Information Pertinente à partir des données issues des réseaux sociaux. En combinant les techniques de Retrieval-Augmented Generation avec l’exploitation de ces sources sociales