Abstract: To solve the problem of inaccurate identification of user demand information caused by the discrepancy of semantic information and one-sided consideration of demand attribute features in the existing research,a method for identification of key demand information oriented to multi-dimensional attribute features was proposed.The comment information was analyzed using natural language processing.The rough number theory was combined to characterize the uncertainty of cognitive information,and user needs were obtained by matching feature words with different lexical properties.Then,the attribute characteristics that affected the identification of key needs information were analyzed from the intrinsic characteristics and external factors respectively.The rough number Decision Making Trial and Evaluation Laboratory (DEMATEL) method was used to analyze the self-organizing relationships within the requirements,and the user attention of the demand was explored,thus the demand weights from internal and external dimensions was determined.Finally,game theory was introduced to determine the comprehensive weights among different dimensional demands to realize the identification of key demand information oriented to multi-dimensional attribute features.The feasibility of the method was verified by taking the recognition process of key demand information in the innovative design of hand drill as an example.

Key words: multi-dimensional attribute features, user demand, critical information identification, semantic information, product innovation design

摘要： 为解决现有研究因语义信息差异性及需求属性特征考量单一化导致的用户需求信息识别不准确的问题,提出一种面向多维属性特征的关键需求信息识别方法。利用自然语言处理对评论信息进行分析,结合粗数理论表征认知信息的不确定性,并通过匹配不同词性的特征词汇获取用户需求;分别从内在特征和外部因素分析影响关键需求信息识别的属性特征,运用粗数决策试验和评价实验法分析需求内部的自组织关系,并探究需求的用户关注度,从内外两个维度确定需求权重。最后,引入博弈论确定需求的综合权重,实现面向多维属性特征的关键需求信息识别。以手电钻创新设计中的关键需求信息识别过程为例,验证了方法的可行性。

关键词: 多维属性特征, 用户需求, 关键信息识别, 语义信息, 产品创新设计