Nowadays the tendency of people to believe in a piece of news that’s coming on social media is very high and during this pandemic, it is more difficult to know if news heard is fake or not. The role of news in our lives has a great impact at present and also in the past years. Especially today, amid the pandemic, social media platform is being used to spread misinformation or the fake news at lightning speed and causes adverse effects in our lives. This approach helps to overcome this challenge and helps recognize or differentiate between true and false news. The data is collected and the content in the data is used for feature extraction using natural learning processing (NLP) by the technique of vectorizer. The extracted features are then classified using the algorithm passive-aggressive classifier a machine learning algorithm, here the input data successively approaches the algorithm and the machine learning algorithm is been upgraded one by one and not using the batch learning where the whole dataset is evaluated in one single step. This algorithm is suitable for huge datasets since this keeps updating the machine learning model at every step. The main challenge of this project is the real-time dataset collection and we are working on it. The output from the machine learning is then updated in IoT implemented NodeMCU an easy open-source platform for IoT application users and it is a hardware module with inbuilt Wi-Fi that is connected to the cloud so that the operators can access it and then using IoT the fake producers get notified as an alert that the news produced from their site is fake.

Keywords : Fake News, vectorizer, Machine learning algorithm, IoT