基于可拓集的企业数据挖掘应用方案初探

Study on enterprise data mining solution based on extension set

数据本身的质量差造成数据挖掘结论的可信度低已经成为影响数据挖掘应用的重要因素,针对不完备数据设计的清洗算法、容忍算法等都不能从根本上解决这个问题．通过深入分析这一矛盾现象的原因,对企业数据建立物元可拓集合,提出基于可拓方法的数据挖掘企业应用方案．以数据挖掘所需的完备数据集做为条件物元,发现数据质量差距,以事元“数据挖掘咨询”促使物元集可拓域的发展变换,推出以数据挖掘应用咨询带动数据质量改进的措施,从而解决了数据质量的矛盾问题,使数据质量不高的企业也可以实施数据挖掘项目,提高信息决策水平．

Data mining needs high quality data while many enterprises have no good enough data to get a credible conclusion, which prevents its implementation. The reasons have been analysed based on Extension Tbeory, and a new method has been by using Matter Element Analysis and extension set, which solves the conflicting problem by data mining consulting. Its application in a web company shows that it has good practicality which can do data mining projects in low quality data enterprises and increase the decision levels.