TOPIC MODELING ANALYSIS OF 2022 NEWS
Topic Modeling Analysis Revealing the Ukrainian War
Project description
The project focuses on analyzing the major topics and trends that defined the news in 2022, specifically through publications from "Le Monde." The initiative involves the creation of a comprehensive, multilingual corpus covering significant periods throughout the year. Utilizing topic modeling techniques, the project aims to uncover and visualize the key themes and their evolution over time.
The process began with team formation and training in GitLab, followed by the establishment of a shared repository for collaborative development. The data collection involved extracting and categorizing articles using specialized scripts and tools like Pathlib and Feedparser.
For analysis, the team used a LDA model and scripts for topic modeling to generate visualizations of the main topics, employing Spacy, Stanza, and Trankit for text processing, and validating results with PyLDAvis. Special attention was given to the French presidential elections, with targeted analyses for the periods before, during, and after the elections.
The final results were presented through accessible visualizations on a dedicated website, along with explanatory articles about the methods and findings. Overall, the project provides an insightful exploration of the news landscape in 2022, highlighting the dynamic nature of media coverage throughout the year. Visit the project site HERE to see the results of our topic modeling investigation.
Discover more about this project and click on the button below to access the GitHub Repository.
Explore More Projects
If you're interested in exploring more projects, please select another project from the dropdown menu.