The Story
After finished the Kaggle’s TalkingData Click Fraud Detection challenge, people were frustrated. As few days before the final submission, there was a public kernel which blended outputs from others and got a nice 0.9812 score. Then people just copied that kernel and submitted to Kaggle. At the end, there were 1xx copies of that kernel on the upper parts of the leaderboard and took the silver medals.
Yes, it sounds like a sad story for people who spent days and nights to build a better model. But just remember one thing, nobody, nobody can ever take away your knowledge for what you have done. #NeverGiveUp
Pingback: Blending, the Dark Side in Data Science Competition ⋆ Code A Star