基于动态置信度优化Apriori算法的旅游市场决策模型

Decision-making Model of Tourism Market Based on Apriori Algorithm with Optimized Dynamic Confidence

针对标准Apriori算法在旅游市场数据规则挖掘的应用中还存在运算时间长、规则挖掘效果不好等问题。本文提出了一种基于动态置信和事务数据库优化的数据挖掘模型,首先采用贝叶斯算法对数据项进行预估,得到动态的置信度,然后采用m-estimate加以优化,从而让该分类能够更为精确地做出其适当的分类,最后用矩阵的形式来代替事务数据库,并通过这一方法来提高算法的运行效率。另外针对这一模型进行了仿真实验,结果发现,本文设计的经过改进的Apriori算法在挖掘市场数据的时候具有更好的性能。

According to the defects of the standard Apriori algorithm such as long operation time and not good effects of rule mining in the application of data rules mining in tourism market, a data mining model is proposed based on dynamic confidence and transaction database optimization. First, estimate the data items by the Bayesian algorithm, and get the dynamic degree of confidence. Then optimize it by the m-estimate, so that the classification can more accurately to make the appropriate classification. Finally, in the form of matrix instead of the transaction database, and improve the operation efficiency of the algorithm. The simulation experiment results show that, the improved Apriori algorithm has better performance in mining market data.