273,601 Turkish news articles categorized into 17 distinct classes, sourced from both print media and digital news websites. The collection spans an eight-year period from 2010 to 2017, provided by the Interpress media monitoring agency.
Use Cases
- Train a multi-class text classifier to assign one of the 17 categories to Turkish news text
- Perform temporal analysis of news trends in Turkey using the 2010-2017 date range
- Develop natural language processing models specifically for the Turkish language using the large corpus of 273,601 articles
- Compare linguistic styles between print media and web-based news sources
Strengths
- Contains 273,601 individual news records in the Turkish language
- Features 17 distinct news categories for classification tasks
- Aggregates data from both physical print media and digital news websites
- Covers a temporal range of eight years from 2010 to 2017