A hybrid computational intelligence approach for efficiently evaluating customer sentiments in E-commerce reviews

The Internet has opened new interesting scenarios in the fields of e-commerce, marketing and on-line transactions. In particular, thanks to mobile technologies, customers can make purchases in a faster and cheaper way than visiting stores, and business companies can increase their sales volume due to a world-wide visibility. Moreover, online trading systems allow customers to gather all the required information about product quality and characteristics, via customer reviews, and make an informed purchase. Due to the fact that these reviews are used to determine the extent of customers acceptance and satisfaction of a product or service, they can affect the future selling performance and market share of a company because they can also be used by companies to determine the success of a product, and predict its demand. As a consequence, tools for efficiently classifying textual customer reviews are becoming a key component of each e-commerce development framework to enable business companies to define the most suitable selling strategies and improve their market capabilities. This paper introduces an innovative framework for efficiently analysing customer sentiments in textual reviews in order to compute their corresponding numerical rating to allow companies to better plan their future business activities. The proposed approach addresses different issues involved in this significant task: the dimension and imprecision of ratings data. As shown in experimental results, the proposed hybrid approach yields better learning performance than other state of the art rating predictors.

I. Introduction

Internet expansion allows companies to improve their selling capabilities on a world-wide visibility and also enables potential buyers to compare and buy products in a much faster and cheaper way than visiting a store. Studies revealed that when it comes to decision making, customers prefer to gather information online than from in-store due to the richness of the information available to them on the Internet [1], [2], [3]. Many major businesses and online traders enable customers to provide online reviews about products that they have purchased. Customers usually provide a written review accompanied by a numerical rating representing their overall satisfaction level for the transaction. Companies use customer feedback to provide potential buyers with information about the quality and user satisfaction for a given product or service. For this reason, reviews have an important impact on businesses performance and they can affect the performance of a company in terms of sales volume and market shares. Opinion mining of customer review ratings is becoming an important task for e-commerce, because companies can extract customer opinions from reviews about specific products in order to determine customer satisfaction based on the degree of positivity of the reviews, and predict demand of current and future products. The data found between the textual reviews and their numerical ratings can be uncertain, and the size of database storing customer reviews could be too big to be accurately analysed by conventional machine learning algorithms, such as neural networks. In order to deal with such large and noisy information, data mining techniques should be efficiently applied to the data in order to reduce the size of the data. As a consequence, only the most important features should be extracted from the textual reviews and used by the system to automatically compute a corresponding rating. The reduced representation of the data is a more suitable and accurate than raw information directly provided by customers after their purchases, and it can be used for performing future marketing activities. As shown in Section II, several intelligent systems have been designed to face this significant challenge, but none of them addressed the rating prediction issue by considering all of the involved difficulties: the dimensionality of data and system accuracy.