Linear Fuzzy Clustering Techniques With Missing Values and Their Application to Local Principal Component Analysis

Honda, Katsuhiro; Ichihashi, Hidetomo

doi:10.1109/tfuzz.2004.825073

Cited by 90 publications

(28 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the lack of unified mathematical structure, there is no exact Fig. 1 Block diagram of a voice conversion system procedure to extract the principal curves although some approaches have been developed [5,8,14]. Figure 3 shows a typical neural network applied to extract the NLPCs.…”

Section: Principal Component Analysis; From Linear To Local Nonlinearmentioning

confidence: 99%

See 1 more Smart Citation

Unaligned training for voice conversion based on a local nonlinear principal component analysis approach

Makki

Hosseini

Seyyedsalehi

et al. 2009

Neural Comput & Applic

View full text Add to dashboard Cite

During the past years, various principal component analysis algorithms have been developed. In this paper, a new approach for local nonlinear principal component analysis is proposed which is applied to capture voice conversion (VC). A new structure of autoassociative neural network is designed which not only performs data partitioning but also extracts nonlinear principal components of the clusters. Performance of the proposed method is evaluated by means of two experiments that illustrate its efficiency; at first, performance of the network is described by means of an artificial dataset and then, the developed method is applied to perform VC.

show abstract

Section: Principal Component Analysis; From Linear To Local Nonlinearmentioning

confidence: 99%

“…An alternative to these two global approaches is LLPCA [5,8,19,20]. In general, LLPCA partitions the data space into a number of nonoverlapping regions and approximates each region by a localized hyperplane.…”

Section: Introductionmentioning

confidence: 99%

Unaligned training for voice conversion based on a local nonlinear principal component analysis approach

Makki

Hosseini

Seyyedsalehi

et al. 2009

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…First one is the extension of fuzzy c-varies clustering algorithm (FCV), use fuzzy clustering and fuzzy covariance matrix with PCA. The second approach combines fuzzy correlation matrix with PCA [16]. Atkinson et al estimated the abnormal data by using EM algorithm or Mutual Information (MI) method [18].…”

Section: Introductionmentioning

confidence: 99%

Fuzzy Classification of Incomplete Data with Adaptive Volume

Yao

Weng

Chang

2009

2009 First Asian Conference on Intelligent Information and Database Systems

View full text Add to dashboard Cite

For solving the incomplete data problem of missing feature values in prototype data, various strategies were proposed. In this paper, two improved approaches are proposed to estimate the missing values of incomplete data. The two approaches are based on combining the adaptive volume GustafsonKessel algorithm (GKA) and the nearest vector features under the distance norm evaluated by complete data. The GKA with adaptive volume is applied for clustering and classifying the results. At last, compared the result with other strategies, and the computer simulations show that the improved strategies provide superior effects.

show abstract

“…When a data set includes large number of objects and the number exceeds the data dimensionality, some method as local principal component analysis might be effective [21]. But, if the data dimension of x is extremely large (Case 1) and/or x includes missing values (Case 2), one may not be able to explicitly partition objects.…”

Section: Introductionmentioning

confidence: 99%

Fuzzy c-Means Classifier for Relational Data

Ichihashi

Honda

Kuramoto

et al. 2007

2007 IEEE Symposium on Computational Intelligence and Data Mining

View full text Add to dashboard Cite

This paper proposes a relational version of the fuzzy c-means (FCM) classifier in which relational data instead of object data are used. The classifier based on the relational clustering is called "relational classifier". The classifier is useful when a feature space has an extremely high dimensionality that exceeds the number of objects and many of the feature values are missing, or when only relational data are available instead of the object data. The relational data is represented by a matrix in terms of distances (dissimilarity) between object data, and is not concerned with the relational database. The clustering algorithm used in the classifier includes, as a special case, the relational dual of FCM proposed by Hathaway, Davenport and Bezdek and can be seen as a simultaneous application of multidimensional scaling and clustering.The computational intensity of the classifier is comparable to Gaussian mixture classifier (GMC). The proposed classifier outperforms well established relational classifier known as knearest neighbor (k-NN) on several benchmark datasets from the UCI ML repository.

show abstract

Linear Fuzzy Clustering Techniques With Missing Values and Their Application to Local Principal Component Analysis

Cited by 90 publications

References 13 publications

Unaligned training for voice conversion based on a local nonlinear principal component analysis approach

Unaligned training for voice conversion based on a local nonlinear principal component analysis approach

Fuzzy Classification of Incomplete Data with Adaptive Volume

Fuzzy c-Means Classifier for Relational Data

Contact Info

Product

Resources

About