A Fusion Approach of Dependency Syntax and Sentiment Polarity for Feature Label Extraction in Commodity Reviews
Jianfei Xu
TL;DR
This paper tackles the problem of extracting meaningful feature labels from large-scale commodity reviews where traditional methods suffer from robustness issues. It introduces a fusion approach that combines dependency parsing with sentiment polarity to form feature tags consisting of a product feature, a sentiment degree, and an evaluation term, guided by rules and a polarity determination mechanism using HowNet and PMI thresholds $\theta_{HowNet}=0.73$ and $\theta_{PIM}=0.50$. The method is validated on 13,218 reviews from JD.com across four product categories, achieving an accuracy around 0.7, recall around 0.8, and an F-score around 0.8, with disharmony in some cases due to syntactic limitations. The work provides a structured framework that can aid consumers and platforms in better summarizing review content for decision support and marketing analytics, while acknowledging limitations related to dictionary dependence and limited feature tag scope and outlining directions for improved semantics and handling of complex sentences.
Abstract
This study analyzes 13,218 product reviews from JD.com, covering four categories: mobile phones, computers, cosmetics, and food. A novel method for feature label extraction is proposed by integrating dependency parsing and sentiment polarity analysis. The proposed method addresses the challenges of low robustness in existing extraction algorithms and significantly enhances extraction accuracy. Experimental results show that the method achieves an accuracy of 0.7, with recall and F-score both stabilizing at 0.8, demonstrating its effectiveness. However, challenges such as dependence on matching dictionaries and the limited scope of extracted feature tags require further investigation in future research.