Context

Collection of chat messages in night urban city between boys and girls.

Content

Data set of messages (more than 1 million of rows) in Russian language from teenager population taken in period from 2012 to 2016 inclusive

Acknowledgements

All personal info in the message' body were taken from public web source, and, though, are free of use.

Inspiration

This dataset can be used to classify chat messages as male / female.

Key objectives

1. Extract phone numbers from messages. All phone numbers are located in Ukraine and belongs to one from next operators

• +380 50
• +380 95
• +380 66
• +380 99
• +380 63
• +380 73
• +380 93
• +380 68
• +380 67
• +380 96
• +380 97
• +380 98

2. Classify chat messages by gender (male/female)