This repository contains solution files for competition of news scraping and classification. The sites lenta.ru, fontanka.ru, asninfo.ru were used as news sources.
The result of the competition: accuracy = 0.9285
place - 9
-
news_kaggle.ipynb : Jupyter Notebook with exploratory data analysis and pipelines
-
parsers/parser_builds.ipynb: Parser of asninfo.ru
-
parsers/parser_fontanka.ipynb: Parser of fontanka.ru
-
parsers/parser_lenta.ipynb: Parser of lenta.ru