Advancing opinion mining with optimised explicit feature extraction in customer reviews

The growing emphasis on opinion mining highlights the significance of analysing customer opinions regarding a service or product and their impact on purchasing decisions. Although identifying pertinent features within customer review analysis is essential for uncovering the expectations, extracting features from unstructured customer review documents presents significant challenges. Current pattern rules are also insufficient for extracting relevant explicit features, alongside linguistic limitations hindering the processing of customer review documents. Thus, this study enhanced the performance of explicit feature extraction from customer reviews by improving heuristic pattern-based rules. The rules comprised 16 newly constructed rules and 25 rules derived from previous studies. Notably, these 16 new rules could extract explicit features frequently overlooked by existing rules in past studies. An enhanced heuristics pattern-based algorithm was also created to identify and extract explicit features using a set of 41 enhanced heuristic pattern-based rules. Although fewer rules were employed for explicit feature extraction compared with previous studies, no optimisation was used for either the rules or the feature extraction process. Consequently, an average precision of 0.93, a recall of 0.90, and an F-measure of 0.91 were obtained across seven datasets from multiple domains. This proposed algorithm exhibited consistent performance and adaptability across various domains, underscoring the effectiveness of the proposed enhanced heuristics pattern-based approach. Overall, the outcomes demonstrated improved explicit feature extraction capabilities, enabling precise identification of specific product attributes or services referenced by customers. This advancement could then facilitate a more profound understanding of customer preferences, pain points, and desires, with significant practical implications for businesses aiming to comprehend and address their customers’ needs.