Python
Aim:
To enhance the assigning accuracy of former methods in fake news detection using advanced methods.
Abstract:
We are in the age of information, everytime we read a piece of information or watch the news on TV, we look for a reliable source. There are so many fake news spread allover the internet and social media. Fake news is misinformation or manipulated news that is spread across the social media with an intention to damage a person, agency and organization. The spread of misinformation in critical situations can cause disasters. Due to the dissemination of fake news, there is need for computational methods to detect them. So, to prevent the harm that can be done using technology, we have implemented Machine Learning algorithms and techniques such as NLTK, LSTM. Our contribution is bifold. First, we must introduce the datasets which contain both fake and real news and conduct various experiments to organize fake news detector.
Synopsis:
The rise in popularity of social media sites like Facebook, Instagram, and Twitter, etc.., Fake news quickly spread across millions of users in a very short period of time. News with certain topics have high probabilities to be classified as fake news. Some authors have high probability to publish fake news and discussed in EDA part. We predict (based on topics).
Proposed System:
We have proposed with the idea of using Bidirectional LSTM with GRU. So the bidirectional LSTM is capable of storing the previous sentence. So we can store the long titles and able to predict with the good accuracy.
Advantage:
Bi-LSTM are much better at handling long-term dependencies compared to RNN. This is due to their ability to remember information for extended periods of time. LSTMs are much less susceptible to the vanishing gradient problem.
Reviews
There are no reviews yet.