计算机工程与应用
計算機工程與應用
계산궤공정여응용
COMPUTER ENGINEERING AND APPLICATIONS
2010年
10期
119-121
,共3页
维数灾难%特征抽取%相似性度量%鲁棒性
維數災難%特徵抽取%相似性度量%魯棒性
유수재난%특정추취%상사성도량%로봉성
curse of dimensionality%feature extraction%similarity metric%robustness
维数灾难是机器学习算法在高维数据上学习经常遇到的难题,基于局部保持的投影方法(Locality Preserving Projection,LPP),可以很好地解决维数灾难难题.然而传统LPP的相似性度量方法对噪音敏感,为此利用鲁棒路径相似的度量方法.提出一种增强的局部保持投影方法.在高维流形数据上的降维实验证实了该方法对噪声和离群点的有效性.
維數災難是機器學習算法在高維數據上學習經常遇到的難題,基于跼部保持的投影方法(Locality Preserving Projection,LPP),可以很好地解決維數災難難題.然而傳統LPP的相似性度量方法對譟音敏感,為此利用魯棒路徑相似的度量方法.提齣一種增彊的跼部保持投影方法.在高維流形數據上的降維實驗證實瞭該方法對譟聲和離群點的有效性.
유수재난시궤기학습산법재고유수거상학습경상우도적난제,기우국부보지적투영방법(Locality Preserving Projection,LPP),가이흔호지해결유수재난난제.연이전통LPP적상사성도량방법대조음민감,위차이용로봉로경상사적도량방법.제출일충증강적국부보지투영방법.재고유류형수거상적강유실험증실료해방법대조성화리군점적유효성.
Curse of dimensionality often comes out when learning from high dimensional dataXocality Preserving Projection(LPP) can preserve the local geometry of the manifold of data and solves this problem well.However,the similarity utilized in LPP is sensitive to outliers and noise,thus an Enhanced Locality Preserving Projection(ELPP) method is proposed with the robust path based similarity.Experiments on high dimensional manifold data prove the effectiveness of this method.
