A Fuzzy Decision Support Method of Product Purchase Based on Text Reviews

Abstract

: Electronic commerce platforms provide decision support for their users due to information overload. How to rank products through online reviews is an important research topic. And, hesitancy and fuzziness exist in text reviews. This paper applies fuzzy logic and methods to solve decision support problems. Given this, the proposed method introduces probabilistic intuitionistic hesitant fuzzy numbers (PIHFNs) to characterize text reviews. A PIHFN can reflect multiple positive, neutral and negative information in text reviews. In addition, this paper proposes a distance formula based on PIHFNs, which is used in the proposed method to provide decision support. The proposed method generates suggested products through the technique for order preference by similarity to an ideal solution. The performance of the method in terms of accuracy is measured by the total relative difference metric. The comparisons and experiments are further conducted to illustrate the advantages of the proposed method. Converting the identified positive, neutral, and negative sentiment orientations into PIHFNs is a new idea for processing online reviews. The proposed method can be applied to tackle the decision support problem accurately.