基于粒计算的增量式知识获取方法精选.pdfVIP

第 33 卷第 2 期 电 子 与 信 息 学 报 Vol.33No.2 2011 年 2 月 Journal of Electronics Information Technology Feb. 2011 基于粒计算的增量式知识获取方法 张清华*①② 幸禹可② 周玉兰② ①(重庆邮电大学数理学院 重庆 400065) ②(重庆邮电大学计算机科学与技术研究所 重庆 400065) 摘 要：该文在研究粒计算理论的基础上，提出了一种基于粒计算的增量式知识获取方法。该方法通过建立决策信 息系统原始的知识粒树，对新增数据，在原始知识粒树中查找相匹配的知识粒，并依据决策值更新知识粒树，实现 快速高效地处理动态信息系统。算法分析及实验对比结果表明，该方法在动态信息系统知识获取方面优于 RGAGC 和 ID4 方法。 关键词：数据挖掘；知识粒树；粒计算；动态信息系统；增量式知识获取 中图分类号：TP301.6 文献标识码： A 文章编号：1009-5896(2011)02-0435-07 DOI: 10.3724/SP.J.1146.2010.00217 The Incremental Knowledge Acquisition Algorithm Based on Granular Computing ①② ② ② Zhang Qing-hua Xing Yu-ke Zhou Yu-lan ① (College of Mathematics Physics, Chongqing University of Posts and Telecommunications, Chongqing 400065, China) ② (Institute of Computer Science Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China) Abstract: A new incremental knowledge acquisition method based on granular computing theory is proposed. First, an original knowledge granule tree is established according to the decision-making information system. Then, for any new additional data, its matched knowledge granule in original knowledge granule tree is found at first, and then the original knowledge granule tree is updated according to the corresponding decision-making value. The new method is an efficient tool for processing dynamic data information. Both algorithm analysis and experiment results show that the new method for processing d