熵值距离的离群点检测及其在学生评教中的应用

Entropy Distance-Based Outlier Detection and Its Application in Student Ratings of Teaching Effectiveness Evaluation

离群数据检测是找出与正常数据不一致的数据。学生评教中由于某种原因,会出现一些评教噪声数据。针对学生评教中噪声数据的特征,提出了一个基于熵值距离的离群点检测算法,该算法通过比较每个数据点所对应的熵值和整个数据集的熵值,来判断数据点的离群程度。仿真结果表明该算法对学生评教中出现的噪声数据具有较好的过滤效果。

Outlier detection is to identify the inconsistent data that is different to the normal data.For some reason,there will be some noise data in the student ratings of teaching evaluation.Based on the characteristics of noise data in teaching evaluation,this paper proposes an entropy distance-based outlier detection algorithm.The algorithm by comparing the entropy between each data corresponding to and the entire data set judges the degree of outlier data.Simulation results show that the algorithm appears to have a good filtering effect to noise data in student ratings of teaching effectiveness evaluation.