 Medical Question Answering using LLMs (Medical RAG) Translating Persian inquiries into English, processing them using pretrained language models to generate accurate responses, and translating the responses back into Persian to enhance precision in medical tasks. |  News Document & Token Classification This project involves classifying documents and tokens, and recognizing named entities. Document classification uses Naive Bayes and Transformers. Token classification uses a text-question input to provide an answer, employing two approaches. Named Entity Recognition identifies and categorizes named entities in unstructured text. |  Illegal Word Recognition Develop a system that can identify altered illicit Persian words, even when interspersed with non-Persian characters, numbers, and special characters, and can also recognize abbreviations and contextually equivalent words. |
 Gender Bias Detection & De-biasing This project focuses on the detection gender biases in diverse language models, including those for multiple languages and multilingual models using machine learning models, specifically Bert and XLM-RoBERTa. A significant aspect of the project is the development of de-biasing techniques specifically for Persian language models. |  Personal Info Hider This project's aim is to enhance privacy by concealing personal information in text files. It uses POS tagging and regex-based solutions to effectively remove identifiable details like names, addresses, URLs, email addresses, and more. |  News Retrieval System An advanced search engine utilizing query expansion, classification, clustering, and link analysis. It uses techniques like Boolean search, TF-IDF, FastText, Transformers, and Elasticsearch models. The system features a user-friendly interface and can be deployed on online servers. |
 Automated Medical Report Generation for Fundus Fluorescein Angiography Images This project presents a new model for automated medical report generation from FFA images. It uses a CNN for visual feature extraction, Cross-modal Memory for feature alignment, and a reinforcement learning algorithm guided by NLG metrics for report generation. | Image Captioning An RNN-based system that generates relevant captions for images. It utilizes pre-trained ResNet50 for image feature extraction and LSTM networks for text generation. |  Machine Learning Classical Model This project involves the implementation of three classical machine learning models: Linear Regression, Decision Tree, and SVM. The models are trained and evaluated on various datasets, with preprocessing tasks such as PCA and EDA also performed. |
 Neural Network From Scratch This project involves the development of a neural network from scratch using NumPy, applied to the CIFAR-10 dataset for image classification. | Content Management System This initiative is a flexible content management platform that enables you to construct and administer various kinds of libraries (such as videos, music, documents, images) and distribute them among your friends. | Web Channel Developed a social networking platform based on Object-Oriented Design and design patterns, where authors and content creators can generate income. This platform allows them to create, share, and monetize their content, fostering a community that rewards creativity. |