Abstract: The paper presents a computational method that mines, processes and analyzes Twitter data for detecting public sentiment of medicine. Self-reported patient data are collected over a period of three months by mining the Twitter feed, resulting in more than 10,000 tweets used in the study. Machine learning algorithms are used for an automatic classification of the public sentiment on selected drugs. Various learning models are compared in the study. This work demonstrates a practical case of utilizing social media in identifying customer opinions and building a drug effectiveness detection system. Our model has been validated on a tweet dataset with a precision of 70.7%. In addition, the study examines the correlation between patient symptoms and their choices for medication.
Keywords: Twitter; social media; data mining; public health