摘要
设计商业智能系统,分别构建ETL模块、数据仓库和OLAP系统。对于Cube中经常出现的异常数据问题,提出使用人工免疫系统进行检测,将Cube查询的KPI历史数据进行二进制编码作为自我集合,并用阴性选择算法产生检测器。设计基于人工免疫检测的新商业智能系统,测试表明,改进的系统可以有效地检测异常数据的存在,从而保障了最终用户端使用数据的准确性和整个商业智能系统的可靠性。
Designed the business intelligence system, and built ETL module, data warehouse and OLAP system. Aiming at the abnormal data in the Cube, introduced the artificial immune system to detect by binary decoding the historical KPI data from Cube as the self set and generating the detectors based on negative selection. Designed the new business intelligence sys- tem using the artifical immune detection. The testing results show that the improved systemcan detect abnormity successfully, which ensures the accuracy of the data that the end customer access and the reliability of the BI system.
出处
《计算机应用研究》
CSCD
北大核心
2010年第1期209-211,共3页
Application Research of Computers
基金
国家自然科学基金资助项目(39880032)
广州市重大科技资助项目(199-2005-001)
广东省自然科学基金资助项目(5004737)
关键词
商业智能
联机分析处理
数据仓库
人工免疫
阴性选择
business intelligence （BI）
on line analytical processing（OLAP）
data warehouse
artifical immunity
negative selection
