复杂网络数据流频繁项集人工智能挖掘仿真

针对传统的复杂网络数据流频繁项集人工智能挖掘方法存在数据挖掘时间较长、准确性较低等问题,提出一种基于时间戳的复杂网络数据流频繁项集人工智能挖掘方法。在训练阶段,利用贝叶斯分类算法找到所有复杂网络数据流频繁项集,并计算不同复杂网络数据流频繁项集的概率估值,在测试阶段,针对不同的测试样本构造不同的分类器,集成分类器,获取分类结果。通过分类结果,构建时间戳的滑动窗口模型,根据滑动窗口的大小对项集进行延迟处理,当项集的类型变化界限超过一定的阈值时,需要重新计算支持度,根据计算结果更新变化界限,完成复杂网络数据流频繁项集人工智能挖掘。实验结果表明,所提方法能够快速、准确地对数据流频繁项集进行人工智能挖掘。

Complex Network Data Flow Frequent Itemset Artificial Intelligence Mining Simulation

Traditionally, the artificial intelligence mining method is easy to lead to long time consumption and low accuracy during the data mining. In this article, an artificial intelligence method to mine the complex network data stream frequent item set based on time stamp is proposed. At the training stage, Bayesian classification algorithm was used to find all frequent item sets of complex network data stream, and the probability estimation values of different frequent item sets of complex network data streams were calculated. At the testing stage, different classifiers were constructed for different test samples. After integrating the classifiers, the classification results were obtained. Through the classification result, the sliding window model of time stamp was constructed, and then the item set was delayed according to the size of sliding window. When the change limit of type of item set exceeded a certain threshold, the support degree needed to be recalculated. According to the calculation result, the change limit was updated. Thus, the artificial intelligence mining of complex network data flow frequent item set was completed. Simulation results show that the proposed method can quickly and accurately perform artificial intelligence mining on frequent item set of data flow.