Fake news analysis modelling
Fake news refers to news containing deceptive or fabricated contents that are actually groundless; they are intentionally overstated or provide false information. This project module proposes a fake news analysis modelling method to identify a variety of features in terms of spreading information.
- Pre-processing of the data using NLP techniques.
- Identification and extraction of novel features that can best capture the deception. Specifically, Use Selenium and Tweepy web scraping tools to collect information of Quote retweets and users respectively. Also, Page rank algorithms to check the credibility of tweets.
- Apply neural network based fake news classifiers
- Apply visualization and statistical analysis techniques to investigate the features. For instance, use propagation of retweets graphs. Test model to successfully classification of tweets into fake and real
The data containing news and veracity information of news can be collected from Kaggle, an open data analysis platform at: https://www.kaggle.com/rmisra/news-category-dataset https://www.kaggle.com/mrisdal/fake-news