TEXT MINING IN BUSINESS. A STUDY OF ROMANIAN CLIENT’S PERCEPTION WITH RESPECT TO USING TELECOMMUNICATION AND ENERGY APPS

Abstract

New technologies have enabled companies to create new and contemporary designs and, at the same time, increase the efficiency of their operations. The Internet has amplified not only the e-commerce with products, but also with services. Paying and changing contracts online in only a few minutes for telecommunication and energy services are only two of the advantages of online APPs. The objective of this study is to show how to use and analyze the review data from customers on online applications or forums, providing very useful insights that could be used to improve company-customer relationships and gain business benefits. The data were collected by using the web scrapping method, obtaining a large amount of text. For the overall understanding of this large amount of text, sentiment analysis was used. Then, to classify the reviews into one of three classes: negative, positive or neutral feeling, five methods were used and compared: logistic regression, decision tree method, K Nearest Neighbors method, SVM algorithm, Naïve Bayes.