基于DM技术动态实现财会软件的智能感知功能

Realizing intellisense function for ledger software using DM technology

介绍了数据挖掘与软件智能感知的基本概念 ,探讨了在软件智能感知实现过程中用数据挖掘技术代替手工处理的可行性 .作者综合知识发现领域已有的研究成果 ,结合财会软件的智能化提出了一个实用的挖掘算法 .它动态地从软件不断积累的数据中提取关联规则知识 ,并利用这些知识根据用户填制凭证时的工作状态智能化地调整科目的显示顺序 .该算法能避免进行大量的运算 。

In order to use data mining technology instead of manual rules collecting method to increase software's intellisense function, the author introduced the elementary concept about data mining and software intellisense, and analyzed the feasibility. Basing on previous research achievements in KDD, proposed a revised association rules mining algorithm to realize ledger software's intellisense function. It could dynamically extract association rules from the gradually accumulated data, and uses them to automatically adjust subject list that presented to users. The algorithm can avoid mass calculation for data mining, and is suitable for real time applications. A real application shows its effectiveness.