基于蚁群-遗传算法的物化视图选取策略  

Solution of Materialized View Selection Based on Ant Colony-Genetic Algorithm

作　　者：龚安[1] 窦万蕊[1] 王彦[2] 

机构地区：[1]中国石油大学(华东)计算机与通信工程学院,东营257061 [2]山东胜利职业学院,东营257022

出　　处：《微计算机应用》2010年第1期15-20,共6页Microcomputer Applications

摘　　要：将蚁群算法和遗传算法相结合用于物化视图选取问题。利用遗传算法较强的全局搜索能力对蚂蚁每次的搜索结果进行优化改良,并在信息素更新时,同时考虑最优、最差路径上的信息素更新。实验结果表明,该算法不仅提高了解的收敛速度,也成功解决了蚁群算法易"早熟"而引起的停滞现象。This paper is intent to unite ant colony algorithm and genetic algorithm for materialized view selection. First, it uses the rapid overall searching ability of genetic algorithm to optimize the searched results of ants during every evolution. Second, the pheromone on the best and on the worst path are both updated. After test, results show it efficiently improved solution convergence rate and successfully resolved the problem that ant colony algorithm falls into local optimal solution easily.

关 键 词：数据仓库 遗传算法 蚁群算法 物化视图选择 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]