数据挖掘技术在消费者偏好中的应用

发布时间：2018-06-21 14:52

本文选题：数据挖掘 + 消费者偏好　；参考：《北京林业大学》2015年硕士论文

【摘要】：市场经济环境下,消费者作为市场经济的核心参与者,对消费者偏好的准确衡量,可以为企业获得利润做出贡献,对其他企业的消费者偏好分析提供借鉴,也将进一步拉动消费,促进社会经济的发展。本文在总结分析消费者偏好相关研究和主要无监督数据挖掘技术的基础上,开展了数据挖掘技术在消费者偏好分析中的应用研究。介于文章使用的数据观测量达到500万以上,并涉及到多个量表,本文特地对数据的基本准备工作进行了阐述,对数据的来源、数据的导入、集成和清洗进行了说明,对不同类型变量的缺失值进行了处理,并从该品牌在全国的门店信息、产品的基本情况和交易情况进行了描述性统计分析,以便更直观的了解数据的基本信息。基于消费者行为学相关理论,在对消费者偏好的实证分析中,用销售总量表征了消费者偏好情况,结合生命周期理论和改进的Bass模型,构建了消费者偏好动态衡量模型,并基于聚类结果进行了偏好的计算,进而说明了消费者对该运动品牌下不同分类产品偏好的差异性。在消费者的购买模式分析中,利用关联规则对交易后台数据进行了分析,得到了消费者在颜色、购买产品类型、购买店铺选择上的偏好情况。总结全文,数据挖掘技术在研究基于数据库提取的消费者偏好中能够进行消费者的分类和购买模式的分析,创新性的基于分类结果构建的消费者动态衡量模型有很强的实用性。
[Abstract]:In the market economy environment, consumers, as the core participants of the market economy, can contribute to the profits of enterprises and provide reference for the analysis of consumer preferences of other enterprises. Will also further pull consumption, promote social and economic development. On the basis of summarizing and analyzing the related research of consumer preference and the main unsupervised data mining technology, this paper develops the application research of data mining technology in consumer preference analysis. In this paper, the basic preparation of the data is expounded, the data source, the data import, the integration and the cleaning are explained. This paper deals with the missing values of different types of variables, and makes a descriptive statistical analysis from the brand's store information in the country, the basic situation of the product and the trading situation, in order to understand the basic information of the data more intuitively. Based on the theory of consumer behavior, in the empirical analysis of consumer preference, the consumer preference is represented by the total sales volume, combined with the life cycle theory and the improved Bass model, the dynamic measurement model of consumer preference is constructed. Based on the clustering results, the preference is calculated, and the difference of consumers' preference for different classification products under the sports brand is explained. In the analysis of consumers' purchase patterns, the transaction background data are analyzed by association rules, and the preferences of consumers in color, product type and shop selection are obtained. In conclusion, the data mining technology can analyze the consumers' classification and purchase patterns in the research of consumer preference based on database extraction, and the innovative dynamic measurement model based on the classification results has strong practicability.
【学位授予单位】：北京林业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F713.55

【参考文献】