Patient opinion mining to analyze drugs satisfaction using supervised learning

Opinion mining is a very challenging problem, since user generated content is described in various complex ways using natural language. In opinion mining, most of the researchers have worked on general domains such as electronic products, movies, and restaurants reviews but not much on health and medical domains. Patients using drugs are often looking for stories from patients like them on the internet which they cannot always find among their friends and family. Few studies investigating the impact of social media on patients have shown that for some health problems, online community support results in a positive effect. The opinion mining method employed in this work focuses on predicting the drug satisfaction level among the other patients who already experienced the effect of a drug. This work aims to apply neural network based methods for opinion mining from social web in health care domain. We have extracted the reviews of two different drugs. Experimental analysis is done to analyze the performance of classification methods on reviews of two different drugs. The results demonstrate that neural network based opinion mining approach outperforms the support vector machine method in terms of precision, recall and f-score. It is also shown that the performance of radial basis function neural network method is superior than probabilistic neural network method in terms of the performance measures used. © 2017 Universidad Nacional Autónoma de México, Centro de Ciencias Aplicadas y Desarrollo Tecnológico. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).