关联规则挖掘在LBS动态数据管理的应用研究

为了满足人们能在任意地点、任意时刻存取任意数据的需求,基于位置的服务(LBS,LocationBasedService)需要进行动态数据管理。一种解决方案是服务器根据关联规则挖掘出的规律,对热点数据进行预测,并利用数据广播技术将热点数据不断地推向移动客户机。但经典的Apriori算法并不适合时序数据的处理,而现有的时序关联规则挖掘算法又对服务的关联时间阀值考虑不够,故本文对经典的Apriori算法进行改进,使之适应动态数据管理的需要,从而为解决LBS动态数据管理问题提出新的解决思路。

The application research of association rule mining on LBS dynamic data management

In order to access and save any data from anywhere at anytime,LBS need to manage data dynamically.One resolution is that server predicts hot data with the rule discovered in association rule mining and uses data broadcasting technology to push hot data to mobile client.But the traditional Apriori algorithm can not deal well with temporal data,and the current algorithms of temporal association rules mining do not take into account the time threshold.So the paper improved the traditional Apriori algorithm to make it fit for dynamic data management.It is a new method to solve the problem of LBS dynamic data management.