Bandile Malaza

Overview

News Sentiment Analyzer is a real-time machine learning web application that fetches live news headlines from a news API, analyzes their sentiment using a fine-tuned transformer model, and displays the results in an interactive dashboard with visualizations. It's designed to help users quickly grasp the tone and emotion of current headlines in an intuitive and visually appealing way.

It fetches live news headlines from the NewsAPI, processes them through a custom machine learning pipeline using a fine-tuned sentiment classification model, and visualizes the results instantly in a live web dashboard built with Streamlit.

🧠 ML Pipeline Overview:
Model: Hugging Face Transformer (cardiffnlp/twitter-roberta-base-sentiment)
Input: Live news headlines from NewsAPI
Preprocessing: Tokenization & batch inference via Transformers
Output: Structured sentiment results → Positive, Negative, or Neutral + confidence scores