机场大数据集离群点检测算法仿真研究

Simulation Research on Outlier Detection Algorithm of Airport Big Data Set

针对传统的离群点检测算法在机场大数据集上存在着检测精度较低、检测运行时间运行较高等问题,提出基于嵌套循环的机场大数据集离群点检测算法。通过对机场大数据集进行分析,采用划分方法对机场数据集进行初步处理,将较大规模的机场数据集划分成小规模的数据子集。利用嵌套循环算法对机场大数据集离群点进行检测,引入两个剪枝规则对其数据集离群点进行高效剪枝,减少了离群点检测时机场数据点之间距离的计算次数,此完成检测。实验结果表明,所提算法具有较高的检测精度并且检测运行时间运行较低。

Traditional outlier detection algorithms have low detection precision and high detection running time for the airport big data set. Therefore, this article proposed an outlier detection algorithm for airport big data set based on nested loop. After analyzing the airport big data set, we used the partitioning method to preliminary process the airport big data set. And then, we divided large-scale airport data sets into small-scale data subsets. Moreover, we used the nested-loop algorithm to detect the outliers in airport big data set. According to the two pruning rules, we pruned the outliers in data set efficiently, reducing the computing times of the distance between the airport data points during the detection of outliers. Thus, we completed the detection. Simulation results show that the proposed algorithm has higher detection accuracy and lower detection running time.