频繁模式挖掘系统的设计与开发  

Design and Implementation of Frequent Pattern Mining System

作　　者：王楠楠 刘慧婷[1] 

机构地区：[1]安徽大学计算机科学与技术学院,安徽合肥230601

出　　处：《计算机技术与发展》2018年第2期150-153,共4页Computer Technology and Development

基　　金：国家自然科学基金(61202227)

摘　　要：在日常生活或者相关科学研究中,使用电子设备会产生大量的数据,如何从数据中删除冗余信息,提取或"挖掘"有用信息就成了当前信息科学和技术领域的一个重要的研究方向。频繁模式挖掘作为众多挖掘算法中的一类基本算法,研究主要包括项目集合、项目序列和时间序列等各种数据中的频繁模式挖掘。频繁模式挖掘算法众多,如数据流频繁闭项集挖掘、不确定数据流的最大频繁项集挖掘和不确定数据的频繁模式匹配。该系统设计的目的是将几个课题组开发的挖掘算法进行集成,并利用可视化界面对算法的性能进行直观的比较。通过系统的可视化界面,可将解决同一问题的多个算法的运行结果放入同一张图中,方便用户查看算法的输出结构并进行算法优劣性的比较。The electronic device can produce a large amount of data in daily life or relevant scientific research. Therefore,how to remove theredundant information,and extract or “mine冶 useful information from the data has become an important research direction presently in thefield of information science and technology. Frequent pattern mining is basic one of many mining algorithms. The research mainly includesthe frequent pattern mining in item sets,item sequence and time series. There are many kinds of frequent pattern mining algorithms like thedata flow frequent closure mining,the maximum frequent item set mining of uncertain data streams,and frequent pattern matching of uncertain data. In order to facilitate users to view the results,we design and implement a system to integrate some pattern mining algorithms together. Through the visual interface of the system,we can put the results of several algorithms which solve the same problem into a diagram,in order to make users view the output structure of the algorithms and compare them.

关 键 词：频繁模式挖掘 可视化 算法集成 算法比较 

分 类 号：TP302[自动化与计算机技术—计算机系统结构]