Classification, Clustering and Dimensionality Reduction

Abstract

Abstract : The primary goal of pattern recognition is supervised or unsupervised classification. Among the various frameworks in which pattern recognition has been traditionally formulated, the statistical approach has been most intensively studied and used in practice. The design of a recognition system requires careful attention to the following issues: feature extraction and selection, cluster analysis, and classifier design and learning. In spite of almost fifty years of research and development in this field, the general problem of recognizing complex patterns with arbitrary orientation, location, and scale remains unsolved. New and emerging applications, such as data mining, web searching, retrieval of multimedia data, face recognition and cursive handwriting recognition, require robust and efficient pattern recognition techniques. The objective of this research proposal is to investigate the following important problems in pattern recognition: (1) classifier evaluation, (2) one-class classification, (3) combination of clustering algorithms, and (4) dimensionality reduction. Solution to these problems will advance the state-of-the-art in pattern recognition, data mining and machine learning. These advances will also be useful to a number of pattern recognition and data mining applications of interest to the Navy.