基于在线点评挖掘的机电产品评价方法研究

发布时间：2018-10-20 11:03

【摘要】：机电产品生产企业保持竞争优势的关键是向用户提供优质、价廉的产品和服务,并不断提升用户的满意度,其关键问题是对产品进行准确的评价,并据此对产品进行有效改进。传统的产品及服务评价一般由领域专家人工进行,但领域专家可能缺乏真实的用户体验,导致其对产品的评价并不被用户真正认可,最终达不到有效提升用户满意度的效果。随着电子商务的迅猛发展,机电产品在线销售已经非常普及。电子商务网站一般都提供了用户在线点评功能。大量用户购买和使用产品后,所填写的评价信息反映了用户真实的产品体验,能够为机电产品提供高效、客观的评价信息。因此本文提出基于在线点评挖掘的机电产品评价方法,以电子商务网站上的用户在线点评数据作为产品评价的主要信息来源,通过对大量点评数据的挖掘来对产品进行客观和真实的评价,以解决专家评价存在缺乏真实的用户体验问题。本文的主要研究内容包括：1)在确定点评数据的来源和分析其特点的基础上,设计面向在线点评的网络爬虫来获取用户点评数据,并对其进行预处理,以满足评价指标体系优化和用户评价识别研究的需求。2)设计面向在线点评的评价指标体系；基于用户知识提出评价指标体系优化方法；基于评价指标体系提出模糊模式匹配模板方法识别在线点评中的用户评价。3)基于模糊综合评价,提出机电产品的评价框架；提出模糊综合评价改进方法以实现从不可信的在线点评数据中获取可信的评价结果。4)针对模糊综合评价方法缺乏对评价结果的具体解释的问题,提出基于模糊认知图的产品个性化评价模型,实现了对评价结果的具体解释,以及更加经济的可替换改进评价指标的推荐,以提升用户满意度。本文提出的基于在线点评的产品评价方法拓展了产品评价的广度和深度,对产品评价有一定的理论指导意义,同时在帮助企业提升产品的用户满意度和市场竞争力方面有一定的实际意义。
[Abstract]:The key to maintain the competitive advantage of mechanical and electrical products manufacturing enterprises is to provide customers with high-quality, inexpensive products and services, and to continuously improve their satisfaction. The key problem is to evaluate the products accurately and improve the products accordingly. Traditional product and service evaluation is generally carried out manually by domain experts, but domain experts may lack real user experience, which leads to product evaluation is not really recognized by users, and ultimately it can not effectively improve the effect of user satisfaction. With the rapid development of e-commerce, mechanical and electrical products online sales have been very popular. E-commerce websites generally provide users with online comment function. After a large number of users purchase and use the products, the evaluation information filled in reflects the real product experience of the users, and can provide efficient and objective evaluation information for mechanical and electrical products. Therefore, this paper puts forward an evaluation method of electromechanical products based on online comment mining, which takes the online comment data of electronic commerce website as the main information source of product evaluation. In order to solve the problem of lack of real user experience in expert evaluation, the product is evaluated objectively and truthfully by mining a lot of comment data. The main research contents of this paper are as follows: 1) on the basis of determining the source and analyzing the characteristics of the data, a web crawler oriented to online review is designed to obtain the data and preprocess it. In order to meet the needs of the optimization of evaluation index system and the research of user evaluation identification. 2) to design an evaluation index system oriented to online evaluation, and to propose an optimization method for evaluation index system based on user knowledge. Based on the evaluation index system, the fuzzy pattern matching template method is proposed to identify the user evaluation in the online evaluation. 3) based on the fuzzy comprehensive evaluation, the evaluation framework of mechanical and electrical products is proposed. An improved fuzzy comprehensive evaluation method is proposed to obtain credible evaluation results from untrusted online evaluation data. 4) in view of the lack of concrete explanation of the fuzzy comprehensive evaluation method, A product individuation evaluation model based on fuzzy cognitive map is proposed, which realizes the concrete explanation of evaluation results and the more economical recommendation of alternative and improved evaluation index, in order to improve user satisfaction. The product evaluation method proposed in this paper based on online evaluation expands the breadth and depth of product evaluation, and has certain theoretical significance for product evaluation. At the same time, it has certain practical significance in helping enterprises to enhance customer satisfaction and market competitiveness.
【学位授予单位】：上海应用技术学院
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F713.36;F407.4

【参考文献】