基于关联规则的超市推荐系统的优化设计

摘要

数据挖掘是指从海量的、无规则的数据中发现潜在的、有用的知识的过程。提出了基于 Apriori 原理的改进算法，主要包括：通过对被扫描数据库事务的缩减来提高算法对频繁项集的挖掘效率；通过优化寻找频繁项集的方法来缩小算法的挖掘时间。对超市的销售记录进行挖掘，找出其中商品的相关性，输入一个用户的购物记录对此用户进行推荐。通过多次实验证实，此算法比传统的算法在寻找全部频繁项集时花费的时间更少。%Data mining refers to discover potentially useful knowledge by processing vast a-mounts of ruleless data.This paper presents an improved algorithm based on the principle of Apriori.The main idea is to scan the database by a reduction of the database transactions and to improve the efficiency of mining frequent itemsets with the algorithm.It can also reduce the time of mining by optimizing the search for frequent itemsets with the algorithm.Through the supermarket's sales records,we can find out the correlation between different commodities and make the user recommendation by entering his/her shopping records.Through many experi-ments the advantage is confirmed that this algorithm is better than traditional algorithm in less time of finding all the frequent item sets.