This is an old revision of the document!

---

Data processing

The data collected was processed in the following steps:

• General cleaning up
• Anonymisation
• Language tagging
• Tokenizing
• Normalization
• PoS tagging