计算机与现代化
計算機與現代化
계산궤여현대화
COMPUTER AND MODERNIZATION
2015年
1期
57-60
,共4页
稀疏贝叶斯回归%残差法%异常检测%回归估计%稀疏性
稀疏貝葉斯迴歸%殘差法%異常檢測%迴歸估計%稀疏性
희소패협사회귀%잔차법%이상검측%회귀고계%희소성
sparse Bayesian regression%residual method%anomaly detection%regress estimation%sparsity
异常检测问题中的数据可以看作是正常信息和异常信息的高度混合,在使得正常信息损失最小的情况下,异常点集合就是前K个包含最多异常信息的样本。启发于这种思想,提出一种基于稀疏贝叶斯回归的异常检测模型,该方法通过在传统的核函数基础上融入Bayesian推理框架,对数据进行回归估计,利用残差法找出偏离程度较大的样本为异常样本。实验结果表明,该方法具有良好的稀疏性和检测精度。
異常檢測問題中的數據可以看作是正常信息和異常信息的高度混閤,在使得正常信息損失最小的情況下,異常點集閤就是前K箇包含最多異常信息的樣本。啟髮于這種思想,提齣一種基于稀疏貝葉斯迴歸的異常檢測模型,該方法通過在傳統的覈函數基礎上融入Bayesian推理框架,對數據進行迴歸估計,利用殘差法找齣偏離程度較大的樣本為異常樣本。實驗結果錶明,該方法具有良好的稀疏性和檢測精度。
이상검측문제중적수거가이간작시정상신식화이상신식적고도혼합,재사득정상신식손실최소적정황하,이상점집합취시전K개포함최다이상신식적양본。계발우저충사상,제출일충기우희소패협사회귀적이상검측모형,해방법통과재전통적핵함수기출상융입Bayesian추리광가,대수거진행회귀고계,이용잔차법조출편리정도교대적양본위이상양본。실험결과표명,해방법구유량호적희소성화검측정도。
The data can be regarded as outliers highly intermixed with normal data in the field of anomaly detection .With mini-mal loss of normal information in the model , outliers are viewed as the top K samples holding maximal abnormal information in a dataset .Inspired by this idea , an anomaly detection model based on sparse bayesian regression which taken the Bayesian inferring framework into traditional kernel function was proposed to find the sample serious deviated from the model though the result of re -gress estimation .Experiment results show that this algorithm is of good sparsity and detection accuracy .