This tool is a news scraper and classifier which is developed for Journalists to have a quick insight about current news. The news article is fetched from multiple pre-defined sources and SVM classifier is used to predict class to which the news article belongs to, further this tool outputs keywords and summary of news article meant for quick insights. This tool uses NLP, Machine learning, Data scraping and Cloud Infrastructure for scaling.
-
Packages:
- pandas
- feedparser
- goose
- RAKE
- matplotlib
- sklearn
- scipy
- numpy
-
Distribution:
- Anaconda2(Optional)
-
Extensions:
- nltk_data