Spectral Analysis of Complex Data

Although data mining has made major advancements since it was first introduced, algorithms capable of handling data with complex characteristics, e.g., high dimensionality, complex graph structures, etc., are only making mainstreams in recent years. Creating such algorithms for complex data presents various challenges, many of which can be overcame by understanding the data characteristics via spectral decomposition. This book aims to study the spectral characteristics of data so that they can be exploited by specific applications to achieve better results. The book will address how spectral information can be integrated into the needs of different analytical tasks and hence, shed some light on this exciting field. The book would be useful to professionals and researchers in data mining and information analysis, or anyone else who are interested in utilizing matrix computation techniques in complex data analysis.