Evaluating User Experiences of Alzheimer’s Drugs from Online Reviews Using Text Mining and Sentiment Analysis

Abstract

Text mining is a new technology that attempts to find useful patterns, trends, patterns and rules from unstructured text data. One of the most commonly used techniques in Text Mining is Sentiment Analysis. Sentiment analysis is the most widely used classification tool to explore an author's attitude. It explores whether the author's attitude is positive, negative or impartial by means of a text. As most of the information in the internet age is found as text, the importance and usage areas of Sentiment analysis are increasing day by day. Sentiment analysis, which is frequently used in social media, can be used to expose users' ideas about a particular topic or product. The aim of this study is to transform drug reviews on websites into meaningful information. This information can help users in decision-making. In this study, personal data obtained from a social platform with Alzheimer's drug reviews of 78 users were evaluated. In particular, the selection of Alzheimer's drugs, unlike other drugs, allows the observations of the patients and relatives of the patient to be evaluated together. The 3723 people who read the review and found it useful strengthens the effect of the comment. In the implementation phase, polarity values of user comments were calculated with Sentiment analysis and Alzheimer's drugs were ranked with the formula developed. In this way, the satisfaction levels of consumers according to the drugs were determined.