基于语义情感分析的电子商务个性化推荐模型研究

发布时间：2018-11-12 14:33

【摘要】：随着Web3.0技术逐渐成熟和社会化媒体技术日益普及,电子商务个性化推荐系统备受关注并广泛应用且呈现社会化、移动化发展趋势,其基于信息推荐机制,利用电子商务网站向客户提供商品的信息和建议,模拟销售人员辅助用户进行商品购买决策。伴随社会化电子商务发展及Web2.0技术发展,商品评论信息日益丰富,但限于用户评分习惯、现有技术局限性,传统基于情感分析的电子商务个性化推荐在推荐精准度、智能性、可扩展性等方面存在局限,严重影响用户使用体验,新出现的语义情感分析技术为解决这些问题提供了可能性,基于此,本文选择基于语义情感分析的电子商务个性化推荐进行研究。本文创新之处包括:(1)将语义情感分析用于电子商务个性化推荐,提高电子商务个性化推荐系统精确度、智能性现有将语义情感分析和电子商务个性化推荐系统作为单独领域,相关研究成果已较多,但将两个领域结合的研究成果则非常少见,故此,本文将两者结合起来构建了基于语义情感分析的电子商务个性化推荐模型,初步揭示了基于语义情感分析的电子商务个性化推荐全貌。(2)提出改进后基于语义情感分析的电子商务用户兴趣建模方法,核心是优化后的用户兴趣相似度算法。本文优化了传统基于情感分析的电子商务个性化推荐方法,主要是用户兴趣建模中兴趣相似度算法。全文共分5章,具体如下:第1章介绍了论文选题背景和研究意义,分析了基于情感分析的信息推荐、电子商务个性化推荐、语义情感分析等主题的国内外研究现状,阐述了论文研究方案、创新点和组织结构等。第2章阐述了电子商务个性化推荐及其典型应用、关键技术等,分析了语义情感分析、基于情感分析的信息推荐等理论及相关技术,为基于语义情感分析的电子商务个性化推荐模型设计及应用案例分析奠定知识基础。第3章根据相关理论及技术,分析设计了基于语义情感分析的电子商务个性化推荐模型的设计目标、遵循的基本原则、设计思路,设计了基于语义情感分析的电子商务个性化推荐模型的体系结构、功能模块、运行机理、技术解决方案。第4章从基于语义情感分析的电子商务个性化推荐技术实现及应用角度进行分析,并阐述了实现模型实现所涉及技术的基础性工作、环境部署、驱动配置,并从所构建的基于语义情感分析的电子商务个性化推荐模型中的四个方面:用户兴趣建模、推荐机制、信息资源管理、语义情感分析,阐述了在餐厅电子商务个性化推荐系统的具体实现,供其他相关应用与实践参考。第5章总结了论文研究工作,展望了后续研究方向。
[Abstract]:With the maturity of Web3.0 technology and the popularity of social media technology, E-commerce personalized recommendation system has been paid attention to and widely used, showing the trend of social and mobile development, which is based on information recommendation mechanism. E-commerce website is used to provide customers with information and advice on products, and to simulate sales personnel to assist users in making purchase decisions. With the development of social electronic commerce and Web2.0 technology, the information of commodity review is more and more abundant, but it is limited to users' scoring habit, the limitation of existing technology, and the traditional recommendation of electronic commerce based on emotion analysis is the accuracy of recommendation. There are limitations in intelligence, scalability and so on, which seriously affect the user's experience. The new semantic emotion analysis technology provides the possibility to solve these problems. In this paper, the semantic emotional analysis based on e-commerce personalized recommendation for research. The innovations of this paper are as follows: (1) the semantic emotional analysis is used in E-commerce personalized recommendation to improve the accuracy of E-commerce personalized recommendation system. Currently, semantic emotional analysis and e-commerce personalized recommendation system are regarded as a separate field in intelligence. There are many related research results, but the research results of combining the two fields are very rare, so, In this paper, a personalized recommendation model of E-commerce based on semantic emotional analysis is constructed by combining the two models. This paper preliminarily reveals the full picture of personalized recommendation of e-commerce based on semantic emotional analysis. (2) an improved modeling method of user interest based on semantic emotional analysis is proposed, the core of which is the optimized similarity algorithm of user interest. This paper optimizes the traditional e-commerce personalized recommendation method based on emotion analysis, which is mainly interest similarity algorithm in user interest modeling. The thesis is divided into five chapters: chapter 1 introduces the background and significance of the thesis, analyzes the research status of information recommendation based on affective analysis, e-commerce personalized recommendation, semantic emotional analysis and other topics at home and abroad. The research scheme, innovation points and organization structure of the thesis are expounded. In chapter 2, the author expatiates on E-commerce personalization recommendation and its typical application, key technology and so on, analyzes the theory and related technology of semantic emotion analysis, information recommendation based on emotion analysis, etc. It lays a knowledge foundation for the design of E-commerce personalized recommendation model based on semantic emotional analysis and application case analysis. In chapter 3, according to the relevant theory and technology, the design goal, the basic principles and the design ideas of the E-commerce personalized recommendation model based on semantic emotional analysis are analyzed and designed. The architecture, function module, operation mechanism and technical solution of the personalized recommendation model of e-commerce based on semantic emotional analysis are designed. Chapter 4 analyzes the realization and application of E-commerce personalized recommendation technology based on semantic emotion analysis, and expounds the basic work, environment deployment, driving configuration of the technology involved in the realization of the model. And from the following four aspects: user interest modeling, recommendation mechanism, information resource management, semantic emotion analysis, This paper expounds the realization of personalized recommendation system in restaurant e-commerce for reference to other related applications and practices. Chapter 5 summarizes the research work of the thesis and looks forward to the future research direction.
【学位授予单位】：湖北工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP391.3

【参考文献】