/Fake-News-Detector

Exploited GoogleBERT embeddings on LIAR dataset for multi-class classification task of Fake news detection.

Primary LanguagePython

Fake-News-Detection

Fake News Detection Survey:

The Research Papers attached at https://github.com/abdullahkhilji/Fake-News-Detection follows the nomenclature of <sr-no>-<yyyy>-<mm>-<original-file-name> and the list below follows - order. (Markers §ø {} used as file references)

Any additions to the files should follow the same standards. § marked files are from the previous Google Doc i.e. Fake News Detection. ø files are included in the repository.

Largely an unsupervised learning problem, since there are very few labelled datasets for supervised learning approach. The unsupervised approach largely takes the problem as an anomaly detection task wherein documents deviating from the general character are labelled as fake, inculcating a general assumption that most of the news are not fake.