Cluster Analysis for Data Mining and System Identification

• Κύριοι συγγραφείς: Abonyi, János (Συγγραφέας), Feil, Balázs (Συγγραφέας)
• Συγγραφή απο Οργανισμό/Αρχή: SpringerLink (Online service)
• Μορφή: Ηλεκτρονική πηγή Ηλ. βιβλίο
• Γλώσσα: English
• Έκδοση: Basel : Birkhäuser Basel, 2007.
• Θέματα: Mathematics.; Applied mathematics.; Engineering mathematics.; Statistics.; Applications of Mathematics.; Statistical Theory and Methods.; Statistics and Computing/Statistics Programs.; Statistics for Business/Economics/Mathematical Finance/Insurance.; Statistics for Engineering, Physics, Computer Science, Chemistry and Earth Sciences.; Statistics for Life Sciences, Medicine, Health Sciences.

 This book presents new approaches to data mining and system identification. Algorithms that can be used for the clustering of data have been overviewed. New techniques and tools are presented for the clustering, classification, regression and visualization of complex datasets. Special attention is given to the analysis of historical process data, tailored algorithms are presented for the data driven modeling of dynamical systems, determining the model order of nonlinear input-output black box models, and the segmentation of multivariate time-series. The main methods and techniques are illustrated through several simulated and real-world applications from data mining and process engineering practice. The book is aimed primarily at practitioners, researches, and professionals in statistics, data mining, business intelligence, and systems engineering, but it is also accessible to graduate and undergraduate students in applied mathematics, computer science, electrical and process engineering. Familiarity with the basics of system identification and fuzzy systems is helpful but not required. Key features: - Detailed overview of the most powerful algorithms and approaches for data mining and system identification is presented. - Extensive references give a good overview of the current state of the application of computational intelligence in data mining and system identification, and suggest further reading for additional research. - Numerous illustrations to facilitate the understanding of ideas and methods presented. - Supporting MATLAB files, available at the website www.fmt.uni-pannon.hu/softcomp create a computational platform for exploration and illustration of many concepts and algorithms presented in the book. 

 Classical Fuzzy Cluster Analysis -- Visualization of the Clustering Results -- Clustering for Fuzzy Model Identification — Regression -- Fuzzy Clustering for System Identification -- Fuzzy Model based Classifiers -- Segmentation of Multivariate Time-series.