Pattern Synthesis for Nonparametric Pattern Recognition

Parametric methods first choose the form of the model or hypotheses and estimates the necessary parameters from the given dataset. The form, which is chosen, based on experience or domain knowledge, often, need not be the same thing as that which actually exists (Duda, Hart & Stork, 2000). Further, apart from being highly error-prone, this type of methods shows very poor adaptability for dynamically changing datasets. On the other hand, non-parametric pattern recogni-tion methods are attractive because they do not derive any model, but works with the given dataset directly. These methods are highly adaptive for dynamically changing datasets. Two widely used non-parametric pattern recognition methods are (a) the nearest neigh-bor based classification and (b) the Parzen-Window based density estimation (Duda, Hart & Stork, 2000). Two major problems in applying the non-parametric methods, especially, with large and high dimensional datasets are (a) the high computational requirements and (b) the curse of dimensionality (Duda, Hart & Stork, 2000). Algorithmic improvements, approximate methods can solve the first problem whereas feature selection (Isabelle Guyon & Andre Elisseeff, 2003), feature extraction (Terabe, Washio, Motoda, Katai & Sawaragi, 2002) and bootstrapping techniques (Efron, 1979; Hamamoto, Uchimura & Tomita, 1997) can tackle the second problem. We propose a novel and unified solution for these problems by deriving a

[1]  M. Narasimha Murty,et al.  An incremental data mining algorithm for compact realization of prototypes , 2001, Pattern Recognit..

[2]  M. Narasimha Murty,et al.  Partition based pattern synthesis technique with efficient algorithms for nearest neighbor classification , 2006, Pattern Recognit. Lett..

[3]  Osamu Katai,et al.  Attribute Generation Based on Association Rules , 2002, Knowledge and Information Systems.

[4]  M. Narasimha Murty,et al.  An incremental prototype set building technique , 2002, Pattern Recognit..

[5]  David G. Stork,et al.  Pattern classification, 2nd Edition , 2000 .

[6]  M. Narasimha Murty,et al.  Fusion of multiple approximate nearest neighbor classifiers for fast and efficient classification , 2004, Inf. Fusion.

[7]  John Wang,et al.  Data Mining: Opportunities and Challenges , 2003 .

[8]  M. Narasimha Murty,et al.  Overlap pattern synthesis with an efficient nearest neighbor classifier , 2005, Pattern Recognit..

[9]  Philip Calvert,et al.  Encyclopedia of Data Warehousing and Mining , 2006 .

[10]  M. Narasimha Murty,et al.  An Efficient Parzen-Window Based Network Intrusion Detector Using a Pattern Synthesis Technique , 2005, PReMI.

[11]  David Hinkley,et al.  Bootstrap Methods: Another Look at the Jackknife , 2008 .

[12]  Doheon Lee,et al.  Improving Classification Accuracy of Decision Trees for Different Abstraction Levels of Data , 2005, Int. J. Data Warehous. Min..

[13]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[14]  John Wang Montclair.  Data Warehousing and Mining : Concepts , Methodologies , Tools , and Applications , 2008 .

[15]  Omar Boussaïd,et al.  Pattern Mining and Clustering on Image Databases , 2009, Database Technologies: Concepts, Methodologies, Tools, and Applications.

[16]  Yoshihiko Hamamoto,et al.  A Bootstrap Technique for Nearest Neighbor Classifier Design , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Petra Perner,et al.  Data Mining - Concepts and Techniques , 2002, Künstliche Intell..

[18]  Qin Ding,et al.  k-nearest Neighbor Classification on Spatial Data Streams Using P-trees , 2002, PAKDD.

[19]  Belur V. Dasarathy,et al.  Data mining tasks and methods: Classification: nearest-neighbor approaches , 2002 .