During the past few years, data mining has grown rapidly in visibility and importance within information processing and decision analysis. This is par ticularly true in the realm of e-commerce, where data mining is moving from a "nice-to-have" to a "must-have" status. In a different though related context, a new computing methodology called granular computing is emerging as a powerful tool for the conception, analysis and design of information/intelligent systems. In essence, data mining deals with summarization of information which is resident in large data sets, while granular computing plays a key role in the summarization process by draw ing together points (objects) which are related through similarity, proximity or functionality. In this perspective, granular computing has a position of centrality in data mining. Another methodology which has high relevance to data mining and plays a central role in this volume is that of rough set theory. Basically, rough set theory may be viewed as a branch of granular computing. However, its applications to data mining have predated that of granular computing.

This book covers methods based on a combination of granular computing, rough sets, and knowledge discovery in data mining (KDD). The discussion of KDD foundations based on the rough set approach and granular computing feature illustrative applications.

This monograph presents novel approaches and new results in fundamentals and applications related to rough sets and granular computing. It includes the application of rough sets to real world problems, such as data mining, decision support and sensor fusion. The relationship of rough sets to other important methods of data analysis – Bayes theorem, neurocomputing and pattern recognition is thoroughly examined. Another issue is the rough set based data analysis, including the study of decision making in conflict situations. Recent engineering applications of rough set theory are given, including a processor architecture organization for fast implementation of basic rough set operations and results concerning advanced image processing for unmanned aerial vehicles. New emerging areas of study and applications are presented as well as a wide spectrum of on-going research, which makes the book valuable to all interested in the field of rough set theory and granular computing.

This book constitutes the refereed proceedings of the Third Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD '99, held in Beijing, China, in April 1999. The 29 revised full papers presented together with 37 short papers were carefully selected from a total of 158 submissions. The book is divided into sections on emerging KDD technology; association rules; feature selection and generation; mining in semi-unstructured data; interestingness, surprisingness, and exceptions; rough sets, fuzzy logic, and neural networks; induction, classification, and clustering; visualization; causal models and graph-based methods; agent-based and distributed data mining; and advanced topics and new methodologies.

This volume contains the papers selected for presentation at the 9th International Conference on Rough Sets, Fuzzy Sets, Data Mining and Granular Computing (RSFDGrC 2003) held at Chongqing University of Posts and Telecommunications, Chongqing, P.R. China, May 26–29, 2003. There were 245 submissions for RSFDGrC 2003 excluding for 2 invited keynote papers and 11 invited plenary papers. Apart from the 13 invited papers, 114 papers were accepted for RSFDGrC 2003 and were included in this volume. The acceptance rate was only 46.5%. These papers were divided into 39 regular oral presentation papers (each allotted 8 pages), 47 short oral presentation papers (each allotted 4 pages) and 28 poster presentation papers (each allotted 4 pages) on the basis of reviewer evaluations. Each paper was reviewed by three referees. The conference is a continuation and expansion of the International Workshops on Rough Set Theory and Applications. In particular, this was the ninth meeting in the series and the first international conference. The aim of RSFDGrC2003 was to bring together researchers from diverse fields of expertise in order to facilitate mutual understanding and cooperation and to help in cooperative work aimed at new hybrid paradigms. It is our great pleasure to dedicate this volume to Prof. Zdzislaw Pawlak, who first introduced the basic ideas and definitions of rough sets theory over 20 years ago.

This book is about Granular Computing (GC) - an emerging conceptual and of information processing. As the name suggests, GC concerns computing paradigm processing of complex information entities - information granules. In essence, information granules arise in the process of abstraction of data and derivation of knowledge from information. Information granules are everywhere. We commonly use granules of time (seconds, months, years). We granulate images; millions of pixels manipulated individually by computers appear to us as granules representing physical objects. In natural language, we operate on the basis of word-granules that become crucial entities used to realize interaction and communication between humans. Intuitively, we sense that information granules are at the heart of all our perceptual activities. In the past, several formal frameworks and tools, geared for processing specific information granules, have been proposed. Interval analysis, rough sets, fuzzy sets have all played important role in knowledge representation and processing. Subsequently, information granulation and information granules arose in numerous application domains. Well-known ideas of rule-based systems dwell inherently on information granules. Qualitative modeling, being one of the leading threads of AI, operates on a level of information granules. Multi-tier architectures and hierarchical systems (such as those encountered in control engineering), planning and scheduling systems all exploit information granularity. We also utilize information granules when it comes to functionality granulation, reusability of information and efficient ways of developing underlying information infrastructures.

Containing twenty six contributions by experts from all over the world, this book presents both research and review material describing the evolution and recent developments of various pattern recognition methodologies, ranging from statistical, linguistic, fuzzy-set-theoretic, neural, evolutionary computing and rough-set-theoretic to hybrid soft computing, with significant real-life applications.Pattern Recognition and Big Data provides state-of-the-art classical and modern approaches to pattern recognition and mining, with extensive real life applications. The book describes efficient soft and robust machine learning algorithms and granular computing techniques for data mining and knowledge discovery; and the issues associated with handling Big Data. Application domains considered include bioinformatics, cognitive machines (or machine mind developments), biometrics, computer vision, the e-nose, remote sensing and social network analysis.