数据挖掘技术在产品配置中的应用研究

发布时间：2018-08-30 14:28

【摘要】：在高速发展的当今社会,人们对产品配置的要求越来越高,传统的配置技术因缺乏配置的柔性,使得配置的质量不能很好的满足用户的需求。随着用户需求的提高,数据库技术也在不断发展壮大,数据挖掘技术在实际应用中已经取得部分成效,论文中提出用数据挖掘技术来辅助产品配置的方法可以很好地弥补传统配置中的不足。用户利用这种产品配置辅助系统可以挖掘出柔性配置规则,结合企业已有的固化知识完成产品配置,并对产品配置结果进行准确评价。论文中设计的产品配置辅助系统是一种松散耦合的系统,该系统包括三个子系统,分别为产品结构配置规则挖掘子系统,产品材料配置规则挖掘子系统及产品配置评价子系统。各个子系统的数据挖掘模型都依企业的应用需求而设计,改进的关联规则算法实现了产品结构配置规则的挖掘,新设计的基于粗糙集的决策规则算法实现了产品材料配置规则的挖掘,最后应用灰色关联度算法实现产品配置的评价。上述三个子系统之间相互独立互不影响,用户在产品配置过程中依据自身需要分别向各个子系统发出任务请求。在每个规则挖掘子系统中挖掘出相应的柔性配置规则,再与系统中已有的领域知识相结合实现产品的灵活配置。在评价子系统中对配置结果进行准确的评价,选择最优的配置编写最终的配置清单。
[Abstract]:With the rapid development of today's society, the requirement of product configuration is more and more high. Because of the lack of flexibility of configuration, the quality of configuration can not meet the needs of users. With the improvement of user demand, database technology is developing and expanding, and data mining technology has achieved some results in practical application. In this paper, the method of using data mining technology to assist product configuration is put forward, which can make up for the deficiency of traditional configuration. The user can mine the flexible configuration rules by using the product configuration assistant system, complete the product configuration with the existing firm knowledge, and evaluate the product configuration result accurately. The product configuration assistant system designed in this paper is a loosely coupled system. The system consists of three subsystems: product configuration rule mining subsystem, product material configuration rule mining subsystem and product configuration evaluation subsystem. The data mining models of each subsystem are designed according to the application requirements of enterprises. The improved association rules algorithm realizes the mining of product structure configuration rules. The newly designed decision rule algorithm based on rough set realizes the mining of product material configuration rules. Finally, the grey correlation degree algorithm is used to realize the evaluation of product configuration. The above three subsystems are independent of each other, and users send out task requests to each subsystem according to their own needs in the process of product configuration. The corresponding flexible configuration rules are mined in each rule mining subsystem, and then the flexible configuration of the product is realized by combining the existing domain knowledge in the system. In the evaluation subsystem, the configuration results are evaluated accurately, and the optimal configuration is chosen to compile the final configuration list.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP311.13

【参考文献】