Two-Way Classification of a Data Table with Non Negative Entries: The Role of the χ2Distance and Correspondence Analysis
- 2 April 2012
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 41 (7) , 1006-1022
Abstract
We developed an approach to two-way classification based on the χ2 distance and Correspondence Analysis. We present, in particular, two classification algorithms: the first one operates a dimension reduction before applying clustering techniques to rows and columns. The second one, successively partitions the data matrix to extract several classification schemes rather than one. Applications to gene expression and web data are presented. The results are compared with an optimal partition algorithm.Keywords
This publication has 25 references indexed in Scilit:
- Latent Block Model for Contingency TableCommunications in Statistics - Theory and Methods, 2010
- Projection-Pursuit Approach for Categorical Data2006
- A Quantitative Study of Gene Regulation Involved in the Immune Response of Anopheline MosquitoesJournal of the American Statistical Association, 2006
- Multinomial-Poisson homogeneous models for contingency tablesThe Annals of Statistics, 2004
- Clustering with block mixture modelsPattern Recognition, 2003
- Bayes FactorsJournal of the American Statistical Association, 1995
- Clustering the rows and columns of a contingency tableJournal of Classification, 1988
- A Criterion for Determining the Number of Groups in a Data Set Using Sum-of-Squares ClusteringBiometrics, 1988
- Estimating the Dimension of a ModelThe Annals of Statistics, 1978
- Hierarchical Grouping to Optimize an Objective FunctionJournal of the American Statistical Association, 1963