Datathon – HackNews – Solution – marmot

Posted Leave a commentPosted in Datathons Solutions

dina zaychik, dzay, denzaymail@gmail.com Sergey Sedov, Sianur, sergey.sedov@gmail.com Task 1. The hypothesis is that propaganda/non-propaganda on article level could be detected using distributional semantics features. That’s why we performed thorough preprocessing, removing urls, hashtags, unusual symbols, unusual articles beginnings, non-English first paragraphs (using langid open package), short texts. After that we trained fasttext supervised model (the […]