OntoOpinionMiner: An Opinion Mining Algorithm for Drug Reviews

The internet is now swamped with user data such as thoughts, comments, reviews, etc. due to the rise of social media in the modern day. This enables the creators to have a deeper understanding of the benefits and drawbacks of their creation totally based on the opinions of end users. Opinion mining has been one of the most studied fields along the lines of data mining and natural language processing in recent years. However, the methods to extract the important parts of the healthcare sector are still lacking in an abundance of literature. Patients constantly look for reviews of a specific drug from other users. In order to estimate the drug satisfaction rate among experienced patients, the proposed approach includes a novel three-fold feature selection and classification architecture. For improved outcomes, the Random-Forest Classifier and Long Short-Term Memory (LSTM) have been combined in the construction of this classifier. The outcomes obtained support the claim that this strategy is superior than traditional individual strategies. The proposed methodology has an average precision of 95.83.