Get the latest tech news

1,600 days of a failed hobby data science project

I dedicated over 1,600 days to a data science project, only to see it fail. In this article, I share a new checklist I developed to help prevent similar setbacks in the future.

My plan was to scrape the entire newsticker over time and analyse it for both metadata patterns, such as posting frequency, and deeper semantic insights about the people, topics and countries involved. Every day featured a list of short news stories (shown in green), each containing a timestamp, headline, body text and optional multimedia elements like Twitter/X feeds, videos or audio clips. Personally, I would have loved to use the beautiful applied ML opportunity of identifying the right newsticker link on the Tagesschau website using NLP to remove any manual labour but unfortunately, I don’t have the time to do so.

Get the Android app