273,601 Turkish news articles categorized into 10 distinct classes sourced from print media and digital news websites between 2010 and 2017. The dataset was curated by Interpress media monitoring and specifically restructured for simplified classification tasks.
Use Cases
- Train a Turkish text classifier to predict one of the 10 news categories based on the article text
- Perform temporal analysis of Turkish news trends using the 2010-2017 timeframe
- Compare linguistic styles between print media and digital news website sources
Strengths
- 273,601 news articles spanning a seven-year period from 2010 to 2017
- Categorized into 10 distinct news classes for simplified multi-class classification
- Aggregated from both traditional print media and digital news websites
- Curated and restructured by Interpress media monitoring company