淮阴师范学院学报(自然科学版）

数字经济背景下产品评论是反映消费者真实需求和情感的重要信息源，也是挖掘和分析产品痛点的重要资源.然而产品评论中大量的噪声信息及消费者情感表达的多样性和复杂性使产品痛点的识别和量化面临挑战.为帮助企业准确定位产品痛点，提高产品竞争力，满足消费者需求，创新产品优化策略，构建了基于观点抽取的产品痛点量化模型.利用依存句法技术，制定了评价对象、情感词、修饰词的提取规则和过滤规则，从产品评论中抽取完整的观点组合；定义了观点组合情感值和评价对象情感值的计算方法，并通过粗粒度属性与细粒度特征之间的对应关系，构建了产品痛点量化模型.实证证明：模型可以有效识别和量化产品痛点，为企业迭代、升级产品，改善用户体验提供科学决策依据.

In the context of the digital economy, product review is an important source of information that reflects the real needs and emotions of consumers, and are also an important resource for exploring and analyzing product pain points. However, the diversity and complexity of a large amount of noise information and consumer emotional expression in product reviews challenge the identification and quantification of product pain points. In order to help enterprises accurately locate product pain points, improve product competitiveness, meet consumer needs, and innovate product optimization strategies, this paper builds a quantitative model of product pain points based on opinions. The model uses dependent syntax technology to formulate extraction rules and filtering rules for evaluation objects, emotional words and modifiers, and extracts a complete combination of opinions from product reviews. This paper defines the calculation method of the emotional value of the point of opinion combination and the emotional value of the evaluation object, and passes the correspondence between coarse-grained attributes and fine-grained features, a quantitative model of product pain points has been built. Empirical evidence proves that the model in this paper can effectively identify and quantify the pain points of the product, provide a scientific decision-making basis for enterprises to iterate and upgrade products and improve user experience.