重庆理工大学学报（自然科学版）
重慶理工大學學報（自然科學版）
중경리공대학학보（자연과학판）
JOURNAL OF CHONGQING INSTITUTE OF TECHNOLOGY
2015年
5期
66-70
,共5页
异常检测%数据挖掘%K-mean 聚类算法%初始聚类中心%加权欧式距离
異常檢測%數據挖掘%K-mean 聚類算法%初始聚類中心%加權歐式距離
이상검측%수거알굴%K-mean 취류산법%초시취류중심%가권구식거리
anomaly detection%data mining%K-means%initial clustering centers%weighted Eucli-dean distance
为提高 K-means 聚类算法在异常检测中的效果，给出一种改进的 K-means 聚类算法。基于最大距离选取初始聚类中心，并引入信息熵计算各个属性的权重，用改进后的加权欧氏距离公式计算数据集中样本点间的距离。选取 KDD CUP99数据集测试算法的性能。实验结果表明，本算法有助于提高异常检测的检测率和降低误报率。
為提高 K-means 聚類算法在異常檢測中的效果，給齣一種改進的 K-means 聚類算法。基于最大距離選取初始聚類中心，併引入信息熵計算各箇屬性的權重，用改進後的加權歐氏距離公式計算數據集中樣本點間的距離。選取 KDD CUP99數據集測試算法的性能。實驗結果錶明，本算法有助于提高異常檢測的檢測率和降低誤報率。
위제고 K-means 취류산법재이상검측중적효과，급출일충개진적 K-means 취류산법。기우최대거리선취초시취류중심，병인입신식적계산각개속성적권중，용개진후적가권구씨거리공식계산수거집중양본점간적거리。선취 KDD CUP99수거집측시산법적성능。실험결과표명，본산법유조우제고이상검측적검측솔화강저오보솔。
In order to increase the performance of the K-means algorithm in anomaly detection,an improved K-means algorithm was proposed. The algorithm selected the initial cluster centers according to maximum distance,and the information entropy was introduced to calculate the weight of every at-tribute,and then the improved weighted Euclidean distance formula was used to calculate the distance between sample points in dataset. The performance of the improved algorithm was tested by KDD CUP99 dataset. The experimental results show that this algorithm will be helpful to increase the detec-tion rate and decrease the false alarm rate in anomaly detection.
