Accurate Fake News Detection in Text-Based Content Using K-Nearest Neighbor, LSTM, MLP and CNN Models

With the proliferation of social media platforms, where users are free to express themselves and share content, the detection of fake news in text has become an important issue. The dissemination of inaccurate or incorrect information is made easier by this, even while it improves communication. The public's capacity to differentiate between genuine content and disinformation is further complicated by the accessibility of diverse internet news sources. Strong methods for detecting and reporting false news must be developed in order to solve this problem. They start by managing missing values, removing noise, tokenising, and stemming the dataset. Our focus is on COVID-19-related fake news in this study. We use the TF-IDF technique to extract features. To improve the identification of false news, we provide CNLSMLKN, a new hybrid model that integrates deep learning and machine learning techniques, including CNN, LSTM, MLP, and KNN. Despite its hybrid architecture's origins in solar irradiance data analysis, it works wonders when it comes to identifying false news. Our findings show that compared to traditional models, the suggested one obtains a far higher prediction accuracy of 98%. The results show that the approach could be useful in the fight against fake news online.