TrendPulse
Real-Time News Data Pipeline
Built a Python data pipeline that collects live trending-news data from an API, cleans and enriches it, performs exploratory analysis, and produces reproducible visual outputs.
System Architecture
API-to-visualization workflow evidenced by the TrendPulse repository outputs.
Live news API
Trending-news responses are captured as reproducible raw data for downstream analysis.
Normalize records
Duplicates, missing values, text fields, and timestamps are standardized.
Feature engineering
Publication-time and title-length features are added to support deeper exploration.
Exploratory statistics
The pipeline summarizes titles, publication patterns, and source frequency.
Saved analytical outputs
Clean CSV data, a written report, and two committed visualizations preserve the results.
A four-stage live-data pipeline with two reproducible visual outputs.
Project Overview
TrendPulse is a Python-based data pipeline that turns real-time trending-news API responses into structured data for analysis and visualization.
The workflow is split into clear stages for collection, preprocessing, exploratory analysis, and visualization. It removes duplicates, handles missing values, and engineers time and text-length features before producing statistical summaries.
The final outputs include cleaned CSV data, a written EDA report, title-length distributions, and source-frequency visualizations, creating a traceable path from raw API data to insights.
Problem
Live news data arrives as changing API responses that must be collected, cleaned, structured, and summarized before trends can be interpreted reliably.
Solution
Created a modular four-stage Python pipeline for API collection, data cleaning, feature engineering, exploratory analysis, and saved visual outputs.