News Article Summarizer Web-App

Overview

The News Article Summarizer and Categorizer is a project aimed at addressing the information overload experienced by users in today's fast-paced world. This project provides a streamlined platform for accessing and summarizing news articles across five main domains: India, World, Business, Technology, and Sports. Additionally, it offers the functionality to convert summarized news articles into audio format, enhancing accessibility and convenience for users who prefer auditory consumption of information.

Features and working

• News article scraping: Collects news articles from Times of India website across various domains.
• Text Summarization: Provides concise summaries of news articles using advanced NLP techniques.
• Text-to-Audio Conversion: Converts summarized news articles into audio format for auditory consumption.

Technologies Used

• Libraries:
  • os.path: For filesystem path operations.
  • csv: For reading and writing CSV files.
  • requests: For sending HTTP requests to fetch web pages.
  • pandas: For data manipulation and analysis.
  • nltk.corpus: For accessing natural language corpora and lexical resources.
  • gtts: For converting text to speech.
  • streamlit: For creating interactive web applications.
  • bs4 (BeautifulSoup): For web scraping HTML and XML documents.
  • newspaper: For web scraping news articles from various sources.

Methodology

1. Web Scraping: Automatically extracts information from websites to retrieve news articles.
2. Text Summarization: Uses models like BART from the transformer library to generate concise summaries of articles.
3. Text-to-Audio Conversion: Utilizes the gTTS (Google Text-to-Speech) module to generate audio summaries from the provided text.

Frontend

Deployed version on Streamlit cloud: https://2minutenews.streamlit.app/