Feature selection for helpfulness prediction of online product reviews: An empirical study

Abstract

Online product reviews underpin nearly all e-shopping activities. The high volume of data, as well as various online review quality, puts growing pressure on automated approaches for informative content prioritization. Despite a substantial body of literature on review helpfulness prediction, the rationale behind specific feature selection is largely under-studied. Also, the current works tend to concentrate on domain- and/or platform-dependent feature curation, lacking wider generalization. Moreover, the issue of result comparability and reproducibility occurs due to frequent data and source code unavailability. This study addresses the gaps through the most comprehensive feature identification, evaluation, and selection. To this end, the 30 most frequently used content-based features are first identified from 149 relevant research papers and grouped into five coherent categories. The features are then selected to perform helpfulness prediction on six domains of the largest publicly available Amazon 5-core dataset. Three scenarios for feature selection are considered: (i) individual features, (ii) features within each category, and (iii) all features. Empirical results demonstrate that semantics plays a dominant role in predicting informative reviews, followed by sentiment, and other features. Finally, feature combination patterns and selection guidelines across domains are summarized to enhance customer experience in today’s prevalent e-commerce environment. The computational framework for helpfulness prediction used in the study have been released to facilitate result comparability and reproducibility.

Profiling Users’ Behavior, and Identifying Important Features of Review “Helpfulness”

“…These techniques for review helpfulness prediction range from simple regression algorithms to complex neural networks [62]- [64]. In addition, feature extraction and selection techniques have also been proposed to model the review helpfulness for different products [65]- [67].…”

An Interactive Network for End-to-End Review Helpfulness Modeling

“…-Recent word embeddings learned from shallow neural networks also show promising performance. Following [11], three types of pretrained embeddings are used. SVM classifiers are then trained on review representa- -Sentiment analysis also shows strengths in modeling helpfulness prediction.…”

“…SVM classifiers are then trained on review representa- -Sentiment analysis also shows strengths in modeling helpfulness prediction. Following [11,64], two finegrained sentiment dictionaries are considered. SVM classifiers are then trained on extracted sentiment features.…”

“…Review texts qualitatively describe reviewers' opinions toward product properties. The textual content contains rich information [16], which is an ideal source [11] for learning helpfulness information. On the other hand, star ratings [40] provide a more straightforward form to quantify reviewers' opinions.…”

Reviewing from a Distance – Uncovering the Negativity Bias of Psychological Distance in Online Word-of-Mouth

“…Using a textual analysis, and specifically the rule-based sentiment analysis VADER (Hutto and Gilbert 2014), allowed us to determine whether the consumption experience was overall negative. VADER, an established method to determine the sentiment valence and sentiment intensity of user-generated texts in online environments, has been successfully used to analyze online reviews (Du et al 2019). Instead of relying on lexicons to determine the sentiment, this method incorporates additional rules to improve its validity.…”