Title: Multiple-valued symbolic data clustering: A model-based approach
Authors: Jose Dias - ISCTE - Instituto Universitario de Lisboa (Portugal) [presenting]
Abstract: Symbolic data analysis (SDA) has been developed as an extension to data analysis that handles more complex data structures. In this general framework, the pair observation/variable is characterized by more than one value: from two (e.g., interval-value data defined by minimum and maximum values) to multiple-valued variables (e.g., frequencies or proportions). Clustering of multiple-valued symbolic data is discussed. We propose a new model-based clustering framework based on Dirichlet distributions that includes mixture of regression/expert models. Results are illustrated with synthetic and demographic (population pyramids) data.