關鍵字查詢 | 類別:會議論文 | | 關鍵字:Clustering for multivariate functional data

[第一頁][上頁]1[次頁][最末頁]目前在第 1 頁 / 共有 02 筆查詢結果
序號 學年期 教師動態
1 105/2 統計系 李百靈 副教授 會議論文 發佈 Clustering for Multivariate Functional Data , [105-2] :Clustering for Multivariate Functional Data會議論文Clustering for Multivariate Functional DataPai-Ling Li; Ling-Cheng Kuocluster analysis;functional principal components analysis;multivariate functional data無We propose a multivariate k-centers functional clustering algorithm for the multivariate functional data. We assume that clusters can be defined via functional principal components subspace projection for each variable. A newly observed subject with multivariate functions is classified into a best-predicted cluster by minimizing a weighted distance measure, which is a weighted sum of discrepancies in observed functions and their corresponding projections onto the subspaces for all variables, among all the clusters. The weight of the proposed algorithm is flexible and can be chosen by the objective of clustering. The proposed method can take the means and modes of variation differentials among groups of each variable into account simultaneously. Numerical performance of the proposed met
2 105/1 統計系 李百靈 副教授 會議論文 發佈 Clustering for multivariate functional data , [105-1] :Clustering for multivariate functional data會議論文Clustering for multivariate functional dataLi, Pai-Ling; Kuo, Ling-ChengBook of Abstracts of COMPSTAT 2016A novel multivariate k-centers functional clustering algorithm for the multivariate functional data is proposed. We assume that clusters can be defined via the functional principal components subspace projection for each variable. A newly observed subject with multivariate functions is classified into a best-predicted cluster by minimizing a weighted distance measure, which is a weighted sum of discrepancies in observed functions and their corresponding projections onto the subspaces for all variables, among all the clusters. The weight of each variable represents the importance of a variable to the cluster information and is determined by the within-variable variation or the between-variable correlations. The proposed method can take the means and modes of variation differentials among groups of each variable into account simultaneous
[第一頁][上頁]1[次頁][最末頁]目前在第 1 頁 / 共有 02 筆查詢結果