In this blog, I’ll walk you through how I built a Python script that scrapes political news from Standard Media Kenya, summarizes each article using NLP, and saves the results to a .txt file.

🛠️ What I Built

I wanted a simple tool to help me stay up to date with Kenyan politics — without manually reading every article.

So I built a Python script that:

Scrapes article links from the Politics section
Downloads and parses the full article using newspaper3k
Summarizes each article into 3 sentences using Sumy (LexRank)
Saves the title, URL, summary, and full text to a .txt file

🧠 Tools and Libraries Used

requests + BeautifulSoup — for scraping
newspaper3k — for extracting full article text
sumy — for summarizing using LexRank
nltk — for tokenization
time — for basic delays (not used for auto-scheduling now)

⚠️ What I Struggled With

🔗 Problem: Many URLs were either incomplete or unusable
✅ Fix: I checked and cleaned URLs early before passing to Article().

📄 Problem: Some articles didn’t load or were empty
✅ Fix: Wrapped downloads in a try-except block to skip broken links safely.

📝 Problem: Summaries were hard to read in the output
✅ Fix: Clean formatting and line breaks made the .txt readable.

📁 Output

The .txt file contains:

✅ Title of each article
✅ Link to the article
✅ 3-sentence summary

Each article is neatly separated with clear headings and lines.

📦 GitHub Repo

🔗 View the full project on GitHub

💡 Final Thoughts

This small but powerful project helped me improve my scraping and NLP skills. If you’re looking to automate news consumption or build a simple NLP tool, this is a great place to start.

📰 Scraping and Summarizing Kenyan Political News with Python