| Middle matching mining algorithm |
| |
| 作者姓名： | 郭平 |
| |
| 作者单位： | GUO Ping 1，CHEN Li 2 1Computer Science & Engineering college of Chongqing University，Chongqing 400044，P. R. China 2Computer Science and Technology School of Sichuan University，Chengdu，Sichuan 610065，P. R. China Received 30 April 2003; revised 29 November 2003 |
| |
| 摘 要： | 1. Introduction With the increase in the amount of information, we feel the pressure from 搕he exploring information? and find that we are in a dilemma of having huge data but with little knowledge 1]. Data mining is an effective method to help us to get rid of the morass, and is a process of inferring knowledge from such huge data. A sequence is an ordered list of transactions (or itemsets), in which each transaction is a set of items. Typically there is a transaction-time associated with ea…
|
Middle matching mining algorithm |
| |
| GUO Ping,CHEN Li Computer Science & Engineering college of Chongqing University,Chongqing,P. R. China Computer Science and Technology School of Sichuan University,Chengdu,Sichuan,P. R. China Received April,revised November.Middle matching mining algorithm[J].Journal of Chongqing University,2003,2(2). |
| |
| Authors: | GUO Ping CHEN Li Computer Science & Engineering college of Chongqing University Chongqing P R China Computer Science and Technology School of Sichuan University Chengdu Sichuan P R China Received April revised November |
| |
| Institution: | 1. Computer Science & Engineering college of Chongqing University, Chongqing 400044, P.R.China
2. Computer Science and Technology School of Sichuan University, Chengdu, Sichuan 610065, P.R.China |
| |
| Abstract: | A new algorithm for fast discovery of sequential patterns to solve the problems of too many candidate sets made by SPADE is presented, which is referred to as middle matching algorithm. Experiments on a large customer transaction database consisting of customer_id, transaction time, and transaction items demonstrate that the proposed algorithm performs better than SPADE attributed to its philosophy to generate a candidate set by matching two sequences in the middle place so as to reduce the number of the candidate sets. |
| |
| Keywords: | middle matching algorithm candidate set SPADE minimum support |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《重庆大学学报(英文版)》浏览原始摘要信息 |
| 点击此处可从《重庆大学学报(英文版)》下载免费的PDF全文 |
