Rahma Salhi

As part of this R&D position, I contributed to two innovative projects: - LLM & Multi-Agent System: Designed and implemented an intelligent system combining Large Language Models (LLMs) and autonomous agents to extract and structure information from heterogeneous PDF files. - SQL-to-Text Generation: Developed a module for the automatic generation of natural language explanations from SQL queries, aiming to make databases more accessible to non-technical users. Technologies: Python, NLP, LLMs, SQL, document processing, data structuring, prompt engineering, multi-agent architecture.

𝐏𝐫𝐨𝐣𝐞𝐜𝐭 𝐓𝐢𝐭𝐥𝐞: Medical Data Modeling & Analysis 𝐏𝐫𝐨𝐣𝐞𝐜𝐭 𝐃𝐞𝐬𝐜𝐫𝐢𝐩𝐭𝐢𝐨𝐧: As an aspiring data scientist passionate about healthcare, I developed a robust Django-based platform for medical data modeling and statistical analysis. This project addressed crucial healthcare needs. 𝐊𝐞𝐲 𝐀𝐜𝐜𝐨𝐦𝐩𝐥𝐢𝐬𝐡𝐦𝐞𝐧𝐭𝐬 𝐚𝐧𝐝 𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬: 1- 𝑷𝒓𝒐𝒋𝒆𝒄𝒕 𝑺𝒄𝒐𝒑𝒆: Collaborated with medical experts to define platform requirements for disease prediction, patient outcomes, and treatment recommendations. 2- 𝑫𝒂𝒕𝒂𝒃𝒂𝒔𝒆 𝑫𝒆𝒔𝒊𝒈𝒏: Created a secure database system for medical data, prioritizing data integrity and privacy compliance. 3- 𝑼𝒔𝒆𝒓 𝑰𝒏𝒕𝒆𝒓𝒇𝒂𝒄𝒆: Designed an intuitive web interface for medical professionals and researchers. 4- 𝑺𝒕𝒂𝒕𝒊𝒔𝒕𝒊𝒄𝒂𝒍 𝑨𝒏𝒂𝒍𝒚𝒔𝒊𝒔: Implemented data analysis algorithms using Python libraries (NumPy, Pandas, sklearn). 5- 𝑴𝒐𝒅𝒆𝒍 𝑫𝒆𝒑𝒍𝒐𝒚𝒎𝒆𝒏𝒕: Integrated machine learning models (scikit-learn, TensorFlow) for real-time predictions. 6- 𝑺𝒆𝒄𝒖𝒓𝒊𝒕𝒚: Ensured data security and healthcare compliance through encryption and user authentication. 7- 𝑷𝒆𝒓𝒇𝒐𝒓𝒎𝒂𝒏𝒄𝒆 𝑶𝒑𝒕𝒊𝒎𝒊𝒛𝒂𝒕𝒊𝒐𝒏: Improved efficiency with database indexing and query enhancements. 𝐑𝐞𝐬𝐮𝐥𝐭𝐬: The platform enabled in-depth statistical analysis, disease prediction, research insights, and ensured healthcare data security and compliance.

- Use predictive modeling, statistics, Machine Learning, and other data analysis techniques to explore, and extract insights from structure and unstructured data. - Apply mathematics to data, perform large scale experimentation and build data driven apps to translate data into intelligence, - Utilize efficient data loading and data augmentation techniques. - Perform tasks such as writing scripts, develop pipelines, SQL queries, etc. - Identify valuable data sources and automate collection processes - Undertake pre-processing of structured and unstructured data - Analyze large amounts of information to discover trends and patterns - Build predictive models and deep-learning or machine-learning algorithms - Combine models through ensemble modeling - Present information using data visualization techniques - Propose solutions and strategies to business challenges - Collaborate with engineering and product development teams

𝐒𝐭𝐮𝐝𝐲, 𝐝𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭, 𝐚𝐜𝐪𝐮𝐢𝐬𝐢𝐭𝐢𝐨𝐧, 𝐢𝐧𝐬𝐭𝐚𝐥𝐥𝐚𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐨𝐩𝐞𝐫𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐚 𝐜𝐮𝐬𝐭𝐨𝐦𝐬 𝐢𝐧𝐟𝐨𝐫𝐦𝐚𝐭𝐢𝐨𝐧 𝐬𝐲𝐬𝐭𝐞𝐦 𝐟𝐨𝐫 𝐭𝐡𝐞 𝐓𝐮𝐧𝐢𝐬𝐢𝐚𝐧 𝐜𝐮𝐬𝐭𝐨𝐦𝐬. - Used critical thinking to break down problems, evaluate solutions and make decisions. - Participate in meetings to verify the specifications in the project specifications - Study and implement the functionalities requested by the customers according to each department - Attending lead meetings to make architectural decisions and write technical guidelines - Worked with customers to understand needs and provide excellent service. - Collaborated with team members to achieve target results.

Stock market prediction has been an active area of research for a long time. The Efficient Market Hypothesis (EMH) states that stock market prices are largely driven by new information and follow a random walk pattern. Though this hypothesis is widely accepted by the research community as a central paradigm governing the markets in general, several people have attempted to extract patterns in the way stock markets behave and respond to external stimuli. My End-Year-Project is moving forward and i have been the supervisor to add sentiment analysis section , we test a hypothesis based on the premise of behavioral economics, that the emotions and moods of individuals affect their decison making process, thus, leading to a direct correlation between ”public sentiment” and ”market sentiment”. We perform sentiment analysis on publicly available Twitter data to find the public mood and the degree of membership into 4 classes - Calm, Happy, Alert and Kind.

comaparative study : Creation of a solution for the prediction of the stock exchange actions of Tunisia Time Series Models : - Statical Model : ARIMA - Seasonal ARIMA - Machine Learning Models : LSTM - FbPROPHET Methodology : CRISP-DM Metrics: MAE - MSE - RMSE

Development of a prototype decision support tool for emotion detection (DEEP LEARNING) - videos ( frames of images ) -speech the data : Multimodal EmotionLines Dataset (MELD) the link : https://affective-meld.github.io/

A major player in the Tunisian economy, the Elfouladh company has not ceased since its creation to provide basic products for the construction, agricultural and industrial sectors. Its contribution is as well on the economy as on the development of the human element which constitutes the base of its influence. Discover the process of the steel products supplied by Elfouladh company such as the reinforcing bar, the merchant iron, the drawn wire and the metallic structures have always been treated and manufactured on the basis of the international standards in force.