Machine Learning Techniques Based on Sentiment Analysis for Drug Recommendation System

Kapil Saxena; Aditya Patel; Sadhna K. Mishra

Abstract

A recommended system can help the user make sense of complex information and make well-considered choices. User-generated material is conveyed in human language in various nuanced ways, making it difficult to provide recommendations based on a study of attitudes. However, there has been a shortage of research devoted to health and medical concerns compared to more mainstream areas like product and media critiques and dining-out recommendations. Sentiment analysis (SA) of the healthcare industry as a whole and patient medication experiences has the potential to yield important information on where to focus efforts to enhance public health. Studies in natural language processing and machine learning are working on performing efficient and trustworthy data SA as more and more people publish their opinions online. In this review, we describe an SA-based drug recommendation system using different machine learning techniques. The review paper provides the SA and its different level, approaches, need, and advantages, the recommendation system, their classification techniques, and the machine learning techniques that help accurately predict sentiments of drug reviews. The primary content of these analyses is a commentary on the efficacy and adverse effects of various medications. It is essential not only to find a technique to measure a treatment's efficiency but also to determine the specific disease in which the medicine has its beneficial, neutral, or detrimental effects.

Abstract

Keywords