Anonymous Spreaders Detection of False Information using Incremental Learning Approach

Considering the amount of false information spread rapidly, identifying the spreaders of false information is significant and has become a need of time. Although some models detect the spreaders of incorrect information, authors found that identifying the anonymous spreaders of false information needs to be paid more attention to and addressed. Therefore, the authors proposed the anonymous spreader detection model that extracts relevant implicit features from the text and user-based features using incremental learning techniques. Authors have developed a novel healthcare dataset with an anonymity parameter and a machine learning-based model to detect and classify the anonymous spreaders of false information. With an incremental learning approach, the Logistic Regression and Content Similarity Measure outperformed other classifiers, and also it was observed that their accuracy increased by 5% and 2%.