Extending associative classifier to detect helpful online reviews with uncertain classes

摘要: While online product reviews are valuable sources of information to facilitate consumers’ purchase decisions, it is deemed meaningful and important distinguish helpful from unhelpful ones for consumers facing huge amounts nowadays. Thus, in light review classification, this paper proposes a novel approach identifying helpfulness. In doing so, Bayesian inference introduced estimate the probabilities belonging respective classes, which differs traditional that only assigns class labels binary manner. Furthermore, an extended fuzzy associative classifier, namely GARCfp, developed train helpfulness classification models based on fuzzily partitioned feature values. Finally, data experiments conducted amazon.com reveal effectiveness proposed approach.