Iran War Media Monitor — Monitoring news coverage and…

Pitch

Description

Comments

About this project

Iran War Media Monitor is dedicated to tracking and analyzing media articles about the US-Israeli conflict with Iran. This project enables users to collect articles, identify key authors and outlets, and perform sentiment analysis to understand public opinion regarding the war. The insights gained can help inform discussions and decisions surrounding this critical issue.

Iran War Media Monitor is a powerful tool designed to track and analyze media coverage concerning the ongoing conflict between Iran, Israel, and the United States. This repository provides a comprehensive approach to understanding the narratives and sentiments surrounding this critical geopolitical issue, making it an essential resource for researchers, analysts, and those interested in the discourse of war.

Key Features

Corpus Creation: Accumulate a rich collection of media articles in English to study how the war is represented in news.
Source Tracking: Identify and monitor the voices of various authors and media outlets, enabling a clear understanding of differing perspectives.
Sentiment Analysis: Gain insights into the articles' sentiments, categorizing them as supportive, neutral, or opposing to the war, allowing for a nuanced view of public opinion.

Data Collection and Processing

The project operates on two primary workflows:

News Article Scraping: The first workflow employs a scraper to gather metadata and articles from predefined RSS feeds. Users have the flexibility to modify the feed source within the rss.py file.
Sentiment Processing: The second workflow analyzes the collected articles for sentiment, updating the database with the relevant insights.

Technology Stack

The repository utilizes MongoDB for storing articles, which can be easily set up using Docker for a seamless experience.

Usage Instructions

Execute rss.py to obtain the latest articles from the designated RSS feed.
Run scraper.py to fetch URLs from Google RSS, subsequently scraping the full articles.
Use sentiment.py for performing sentiment analysis on the gathered articles, with a specific focus on identifying their stance regarding the conflict.

Future Enhancements

Future iterations of this project aim to:

Expand the capability to scrape additional RSS feeds to broaden the media coverage.
Develop a custom model for filtering relevant news articles effectively.
Improve the tonality prediction model to enhance categorization of articles as pro, neutral, or opposed to the war.
Incorporate entity and relationship extraction to map out key actors and events.
Apply Generative AI techniques for a comprehensive analysis of the articles and the overall corpus.

This project not only serves as an analytical tool but also fosters a deeper understanding of media influence in war narratives, inviting contributions and collaboration from the community.