Falsified News Detection Using Deep Learning Approach

Abstract

Technology is changing the way that we live our lives. With the advances in technology, Social Media attains a lot of attention from the people around us. With the advancement of technology, fake news with both political or commercial motives is growing very fast on the internet. Users might be affected easily by falsified news, which has brought about mammoth effects on humanity. The foremost goal is to stop the spread of rumors and focus on the true, substantiate news columns. This paper focuses on techniques such as training deep learning neural network models and natural language processing (NLP) techniques for text analytics based on news titles or news content. These techniques can be used for the treatment of the data. Some of the treating steps are tokenizing the input, lemmatizing it, Stemming and removal of irrelevant words (stop words) are done prior to converting them into N-gram vectors using a technique known as TF-IDF that is term frequency-inverse document frequency. With the help of Machine learning and natural language processing, we will try to aggregate the news and later determine whether the news is real or fake using deep neural networks. Experimental evaluation yields the simplest performance using Term Frequency-Inverted Document Frequency (TF-IDF) as a feature extraction technique.