vudaotuan / avito-duplicates-kaggle

Solution for Avito Duplicate Ads Detection competition

Avito Duplicate Ads Detection on Kaggle

A part of the 5th place solution for the ADAD competition run on kaggle

• High level overview of features: https://www.kaggle.com/c/avito-duplicate-ads-detection/forums/t/22190/solution-sharing/126772#post126772
• Slides: http://www.slideshare.net/AlexeyGrigorev/avito-duplicate-ads-detection-kaggle
• This is a code dump with no cleaning. Please let me know if some stuff doesn't work or something is not clear

Running it:

• First all prepare_ and mongo_ files should be run, and also the classes from the java package
• Then all calculate_ files should be run
• Finally, run model_ and stacking at the end