Securing Social Platform from Misinformation Threats Using Deep Learning

Misinformation or fake news spreads much faster than real news on social media that can quickly create panic situations. People are drawn to social media to communicate with one another as technology advances. Furthermore, people are easily duped by fake news and begin spreading it to their networks. During the COVID-19 pandemic, the number of fake news stories on social media increased, confusing people worldwide. Social people can report abusive and unauthentic news posted on the platform. Most of the time, fake news causes panic and forces people to engage in unethical behaviour such as strikes, roadblocks, and other similar actions. Thus, the quality assessment and detection of this fake news must be needed so that people can be safe from misinformation threats on social platforms. Also, quality assessment of posts/news on social platforms must improve the security and privacy of humans. Filtering and quality assessment of fake news manually from many news posts are nearly impossible, raising security and privacy concerns for users. As a result, it is critical to assess the quality of news early on and prevent it from spreading to protect people’s privacy. To address this issue, an automated model is developed in this study to assess the fake news at an early stage that also maintains the quality of the information. The proposed model used a deep learning-based Long-Short Term Memory network with word embedding and outperformed the existing models by achieving 99.82%