News Topic-Modelling Pipeline, Evaluation

Overview

A clone of the open-source Newspaper-Topic-Modelling project, a Python NLP pipeline that ingests news articles, applies topic modelling, and surfaces results through a React frontend. Retained for evaluation; the repository carries upstream history (CZboop) with no studio-authored changes.

Why It Exists

To evaluate an end-to-end topic-modelling workflow over news text, data processing, model training, and a web visualization layer, including its testing and CI setup.

What We Built

No custom development; an upstream snapshot studied as reference. Observed: a Python data-processing pipeline with OS-agnostic path handling and unit tests, a React frontend, and a GitHub Actions workflow deploying both the Python and React jobs.

Technologies & Approach

Python for NLP/topic modelling with a React visualization frontend, wired into CI via GitHub Actions. Reviewed for its pipeline structure and test discipline.

Outcome / Impact

Provided the studio with a working reference for NLP topic-modelling pipelines and their visualization/CI patterns. Documented as evaluation/R&D.

Capabilities Demonstrated

• Evaluating NLP topic-modelling pipelines
• Studying data-processing + visualization + CI workflows