Data Science
Personal Finance Analytics
Statement parsing, auto-categorisation and spend forecasting
- Python
- Pandas
- Prophet
- Streamlit
- +2
Data Science · Mini Project
Pulls comments via the YouTube Data API and charts audience mood
Overview
Paste a video link and the app fetches its public comments through the official YouTube Data API v3, paging through results within the free quota. Comments are cleaned, language-filtered and scored with a pretrained multilingual sentiment model, then shown in a Streamlit dashboard with sentiment split, a timeline, top keywords and the most-liked positive and negative comments.
What makes it stand up
Module breakdown
Fetches comment threads and replies through the API.
Strips links, repeated characters and obvious spam.
Runs batched sentiment inference and stores results.
Extracts frequent terms per sentiment class.
Streamlit charts, filters and CSV download.
After this, you will be able to
Technology stack
You receive
Keep looking
Data Science
Statement parsing, auto-categorisation and spend forecasting