20 Newsgroups NLP Analysis with BERT, BART, and RoBERTa

This repository offers a Flask web application designed for in-depth analysis of the 20 Newsgroups dataset. Utilizing state-of-the-art models like BERT, BART, and RoBERTa, users can classify articles, obtain summaries, and get answers to their questions.

Features:

• Article Classification with BERT: Efficiently categorize articles into predefined topics.
• Document Summarization with BART: Generate concise and coherent summaries of lengthy articles.
• Question Answering with RoBERTa: Extract specific information from articles by posing questions. Two RoBERTa variants are available for this task:
  • consciousAI/question-answering-roberta-base-s-v2
  • deepset/roberta-base-squad2

Technical Details:

• Flask Framework: The application is built on Flask, offering a lightweight web server to interact with the models.
• BERT for Classification: BERT's bidirectional context capturing capabilities are used for classifying articles.
• BART for Summarization: BART, a sequence-to-sequence model, is employed to condense articles into shorter summaries.

Getting Started:

Dataset:

You can download the 20 Newsgroups dataset from this link.

Prerequisites:

• Python 3.x
• Flask
• PyTorch
• Transformers library

Installation:

Clone the repository:

git clone https://github.com/OthmanMohammad/20Newsgroups-QuestionAnswering-Summarization-BERT.git

Navigate to the project directory and install the required packages:

pip install -r requirements.txt

Running the Application:

Start the Flask server:

python app.py

Demo Video

Watch the Demo Video Here