Adaptive Lotus Effect Optimization with DKN for Fake News Detection on Social Media with Tamil Language

Nowadays, people are widely using social media to consume news faster but with the development of various social media platforms, the fake news are spreading faster. Fake news through social media can make public to believe fake stories, it may cause panic and influences the public opinion. Thus, the fake news identification is essential to predict fake news or aiming to flag the specific content. In order to reduce the spread of fake news, this research proposed an effective method to detect fake news using Adaptive Lotus Effect optimization with Deep Kronecker Network (ALEO-DKN). However, the tokenization is done by using Bidirectional Encoder Representations from Transformers (BERT) technique and the features extracted are Word2vec, Number of numerical values, Hashtag, Punctuation marks, Numerical words, Lin similarity, and Term Frequency-Inverse Document Frequency (TF-IDF). At last, fake news identification is done using Deep Kronecker Network (DKN), whereas the introduced Adaptive Lotus Effect optimization (ALEO) is used to optimize the DKN. Meanwhile, the developed model obtained the highest value of precision as 90.8%, Recall as 94.7%, F-measure as 92.7% and rouge as 86.5%.