NEWSMO: A Computer-Aided News Classification Model from Tweets Topic Using a Modified Passive Aggressive and Pipeline Classifier

In the current technology, wherein the internet is wall-to-wall, all of us are predicated on diverse online assets for information. Fake information in recent times is a crucial factor in the existence of social media and the political world. Fake information detection is a crucial study to be carried out for its detection; however, it also has a few demanding situations. Some demanding situations may be because of a much smaller range of assets, such as finding datasets and posted literature. This paper's findings show that the proposed model has achieved various accuracies and comparisons on many machine learning algorithms. The proposed model goal is to allow the user to determine if the information is authentic or fraudulent. Due to a lack of supporting data or assertions, fake news frequently relates to recent, urgent occurrences that existing knowledge sources may not have fully confirmed. The proposed model has utilized a publicly available Kaggle and UCI dataset and modified it to match the current data. It comprises 1856 articles divided into six categories: State, satire, bias, conspiracy, hate, and fake science. The dataset is divided into 371 records for testing and 1485 records for training. The model included modified passive aggressive and pipeline classifiers with 99.6% and 98.9% accuracy, respectively. The objective is to create a system that can correctly categorize news stories that have never been viewed.