Sentiment Analysis: Relationship Between Customer Sentiment and Online Customer Ratings for Price Comparison Engines. An Empirical Study

Sentiment analysis, in other words, opinion mining, is aiming at analyzing people’s sentiments, opinions, emotions, attitudes, etc. Customer sentiment refers to the emotions expressed by customers through their text reviews. These sentiments can be positive, negative or neutral. This study will explore customer sentiments and express them in terms of customer sentiment polarity. In the current days, Greece is facing one of the worst economic crisis in its history, so price comparison engine usage is more than needed, especially for the most competitive and pricy goods, such as athletic footwear and technology ones. In such circumstances, companies have to find more efficient ways to get the absolutely necessary information from their targeted audience by overcoming the problems that a researcher can face with the usage of an ordinary questionnaire, because the customer has written his own point of view; using his own words without being guided by a questionnaire or an interview. This study tries to identify this crucial information and help the contemporary e-shop to improve its ecommerce services and gain more income with less advertising, cpc campaigns, etc. Hence, in this case we gathered from «Skroutz» one of the most renowned PCE in Greece and extracted the sentiment from these core industries target groups, based on the user’s/buyer’s comments and their rating. We used WEKA for classifying the text and extracting knowledge.