Portfolio
Filter by Categories
Système de Détection de Fraude en Temps Réel
Mar,2024
Système de détection de fraude en temps réel avec Apache Spark et modèles de machine learning. Pipelines ETL et modèles (Random Forest, XGBoost, Isolation Forest) pour la détection d'anomalies. Déploiement des modèles en temps réel via MLflow. Tableaux de bord Power BI/Tableau pour le suivi des détections et alertes.
See More
Application d'Aide à l'Obtention de Subventions Utilisant l'IA
Feb,2024
Développement d'une application web IA pour simplifier la gestion des subventions. Intégration de LLMs pour la génération automatique de contenu et analyse de données. Automatisation de la collecte de données via API pour une précision accrue. Gestion de projet en méthode Kanban pour assurer la coordination et livraison des jalons.
See More
Movie Similarity Prediction Using NLP and K-Means Clustering
Jan,2024
In this project, I used Natural Language Processing (NLP) and K-Means Clustering to predict movie similarities. By analyzing the plot summaries and other textual data of movies, the model clusters similar movies together, helping users find movies with related content and themes.
See More
Twitter Sentiment Analysis Model Using NLP and Machine Learning
Jan,2024
In this project, I built a Twitter sentiment analysis model using Natural Language Processing (NLP) and machine learning techniques. The model analyzes tweets to determine the sentiment (positive, negative, or neutral), providing insights into public opinions and trends on various topics.
See More
Real-time person detection and face identification system using YOLO and CNN
Dec,2024
In this project, I collected facial data of six actors from various scenes in a TV show. Using YOLO (You Only Look Once) to detect the person, I then extracted the face from the detected region. Finally, I identified the actors using a Convolutional Neural Network (CNN) model for face recognition.
See More
Clustering for Effective Customer Segmentation and Dataset Analysis
June,2023
In this project, I began by working with a customer selling dataset. After cleaning the data, my focus shifted towards clustering the clients using various clustering algorithms. The goal was to group similar clients together based on their purchasing patterns and behaviors. By applying these clustering techniques, I aimed to uncover hidden information and gain insights into customer segmentation and preferences within the dataset. This analysis provides valuable knowledge that can be utilized for targeted marketing strategies and personalized customer experiences.
See More
Colon Disease Detection with PySpark MLlib
2023
This project aimed to detect colon diseases using PySpark MLlib for processing and analyzing large medical datasets. Steps included data collection, cleaning, and visualization to identify key trends. Machine learning models were then trained to predict the presence of colon diseases, providing accurate results and enabling clear visualizations to enhance data understanding.
See More
Facial-Based Age and Gender Detection
April,2023
This project involved collecting and cleaning a dataset of over 25,000 facial images. We conducted experiments with different models to achieve optimal results for facial analysis tasks such as age and gender detection. Finally, we successfully implemented the chosen model into a web application for practical usage.
See More
Simple Image Processing Desktop Application
April,2022
This desktop application is specifically designed for image processing tasks, allowing users to apply various filters and perform basic editing operations such as rotation, cropping, and selection. It was developed using pyQT, a Python framework for creating graphical user interfaces, and leverages the powerful filtering functions provided by OpenCV. With this application, users can enhance and manipulate images by applying a wide range of filters available in OpenCV, resulting in visually appealing and customized outputs.
See More
About this Page
The website was developed using Next.js and Tailwind CSS by Ayoub Boudra. It is hosted on Vercel.
Get in touch
© 2024 All rights reserved.
Designed & Coded by Ayoub boudra.