This book presents recent developments and research trends in the field of feature selection for data and pattern recognition, highlighting a number of latest advances. The field of feature selection is evolving constantly, providing numerous new algorithms, new solutions, and new applications. Some of the advances presented focus on theoretical approaches, introducing novel propositions highlighting and discussing properties of objects, and analysing the intricacies of processes and bounds on computational complexity, while others are dedicated to the specific requirements of application domains or the particularities of tasks waiting to be solved or improved. Divided into four parts - nature and representation of data; ranking and exploration of features; image, shape, motion, and audio detection and recognition; decision support systems, it is of great interest to a large section of researchers including students, professors and practitioners.

In this book, the following three approaches to data analysis are presented: - Test Theory, founded by Sergei V. Yablonskii (1924-1998); the first publications appeared in 1955 and 1958, - Rough Sets, founded by Zdzislaw I. Pawlak (1926-2006); the first publications appeared in 1981 and 1982, - Logical Analysis of Data, founded by Peter L. Hammer (1936-2006); the first publications appeared in 1986 and 1988. These three approaches have much in common, but researchers active in one of these areas often have a limited knowledge about the results and methods developed in the other two. On the other hand, each of the approaches shows some originality and we believe that the exchange of knowledge can stimulate further development of each of them. This can lead to new theoretical results and real-life applications and, in particular, new results based on combination of these three data analysis approaches can be expected. - Logical Analysis of Data, founded by Peter L. Hammer (1936-2006); the first publications appeared in 1986 and 1988. These three approaches have much in common, but researchers active in one of these areas often have a limited knowledge about the results and methods developed in the other two. On the other hand, each of the approaches shows some originality and we believe that the exchange of knowledge can stimulate further development of each of them. This can lead to new theoretical results and real-life applications and, in particular, new results based on combination of these three data analysis approaches can be expected. These three approaches have much in common, but researchers active in one of these areas often have a limited knowledge about the results and methods developed in the other two. On the other hand, each of the approaches shows some originality and we believe that the exchange of knowledge can stimulate further development of each of them. This can lead to new theoretical results and real-life applications and, in particular, new results based on combination of these three data analysis approaches can be expected.

Decision trees and decision rule systems are widely used in different applications as algorithms for problem solving, as predictors, and as a way for knowledge representation. Reducts play key role in the problem of attribute (feature) selection. The aims of this book are (i) the consideration of the sets of decision trees, rules and reducts; (ii) study of relationships among these objects; (iii) design of algorithms for construction of trees, rules and reducts; and (iv) obtaining bounds on their complexity. Applications for supervised machine learning, discrete optimization, analysis of acyclic programs, fault diagnosis, and pattern recognition are considered also. This is a mixture of research monograph and lecture notes. It contains many unpublished results. However, proofs are carefully selected to be understandable for students. The results considered in this book can be useful for researchers in machine learning, data mining and knowledge discovery, especially for those who are working in rough set theory, test theory and logical analysis of data. The book can be used in the creation of courses for graduate students.

In this book, the following three approaches to data analysis are presented:

- Test Theory, founded by Sergei V. Yablonskii (1924-1998); the first publications appeared in 1955 and 1958,

- Rough Sets, founded by Zdzis aw I. Pawlak (1926-2006); the first publications appeared in 1981 and 1982,

- Logical Analysis of Data, founded by Peter L. Hammer (1936-2006); the first publications appeared in 1986 and 1988.

These three approaches have much in common, but researchers active in one of these areas often have a limited knowledge about the results and methods developed in the other two. On the other hand, each of the approaches shows some originality and we believe that the exchange of knowledge can stimulate further development of each of them. This can lead to new theoretical results and real-life applications and, in particular, new results based on combination of these three data analysis approaches can be expected.

- Logical Analysis of Data, founded by Peter L. Hammer (1936-2006); the first publications appeared in 1986 and 1988.

These three approaches have much in common, but researchers active in one of these areas often have a limited knowledge about the results and methods developed in the other two. On the other hand, each of the approaches shows some originality and we believe that the exchange of knowledge can stimulate further development of each of them. This can lead to new theoretical results and real-life applications and, in particular, new results based on combination of these three data analysis approaches can be expected.

These three approaches have much in common, but researchers active in one of these areas often have a limited knowledge about the results and methods developed in the other two. On the other hand, each of the approaches shows some originality and we believe that the exchange of knowledge can stimulate further development of each of them. This can lead to new theoretical results and real-life applications and, in particular, new results based on combination of these three data analysis approaches can be expected."

Decision trees and decision rule systems are widely used in different applications as algorithms for problem solving, as predictors, and as a way for knowledge representation. Reducts play key role in the problem of attribute (feature) selection. The aims of this book are (i) the consideration of the sets of decision trees, rules and reducts; (ii) study of relationships among these objects; (iii) design of algorithms for construction of trees, rules and reducts; and (iv) obtaining bounds on their complexity. Applications for supervised machine learning, discrete optimization, analysis of acyclic programs, fault diagnosis, and pattern recognition are considered also. This is a mixture of research monograph and lecture notes. It contains many unpublished results. However, proofs are carefully selected to be understandable for students. The results considered in this book can be useful for researchers in machine learning, data mining and knowledge discovery, especially for those who are working in rough set theory, test theory and logical analysis of data. The book can be used in the creation of courses for graduate students.

This monograph is devoted to theoretical and experimental study of partial reductsandpartialdecisionrulesonthebasisofthestudyofpartialcovers. The use of partial (approximate) reducts and decision rules instead of exact ones allowsustoobtainmorecompactdescriptionofknowledgecontainedindecision tables, andtodesignmorepreciseclassi?ers. Weconsideralgorithmsforconstructionofpartialreductsandpartialdecision rules, boundsonminimalcomplexityofpartialreductsanddecisionrules, and algorithms for construction of the set of all partial reducts and the set of all irreducible partial decision rules. We discuss results of numerous experiments with randomly generated and real-life decision tables. These results show that partial reducts and decision rules can be used in data mining and knowledge discoverybothforknowledgerepresentationandforprediction. Theresultsobtainedinthe monographcanbe usefulforresearchersinsuch areasasmachinelearning, dataminingandknowledgediscovery, especiallyfor thosewhoareworkinginroughsettheory, testtheoryandLAD(LogicalAnalysis ofData). The monographcan be usedunder the creationofcoursesforgraduates- dentsandforPh. D. studies. An essential part of software used in experiments will be accessible soon in RSES-RoughSetExplorationSystem(InstituteofMathematics, WarsawU- versity, headofproject-ProfessorAndrzejSkowron). We are greatly indebted to Professor Andrzej Skowron for stimulated d- cussionsand varioussupportof ourwork. We aregratefulto ProfessorJanusz Kacprzykforhelpfulsuggestions. Sosnowiec, Poland MikhailJu. Moshkov April2008 MarcinPiliszczuk BeataZielosko Contents Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 PartialCovers, ReductsandDecisionRules . . . . . . . . . . . . . . . . 7 1. 1 PartialCovers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1. 1. 1 MainNotions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1. 1. 2 Known Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1. 1. 3 PolynomialApproximateAlgorithms. . . . . . . . . . . . . . . . . . 10 1. 1. 4 Bounds on C (?)Based on Information about min GreedyAlgorithm Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1. 1. 5 UpperBoundon C (?). . . . . . . . . . . . . . . . . . . . . . . . . . 17 greedy 1. 1. 6 Covers fortheMostPartofSetCoverProblems. . . . . . . . 18 1. 2 PartialTests and Reducts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 1. 2. 1 MainNotions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 1. 2. 2Relationships betweenPartialCovers and Partial Tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1. 2. 3 PrecisionofGreedyAlgorithm. . . . . . . . . . . . . . . . . . . . . . . 24 1. 2. 4 PolynomialApproximateAlgorithms. . . . . . . . . . . . . . . . . . 25 1. 2. 5 Bounds on R (?)Based on Information about min GreedyAlgorithm Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 1. 2. 6 UpperBoundon R (?). . . . . . . . . . . . . . . . . . . . . . . . . . 28 greedy 1. 2. 7 Tests fortheMostPartofBinaryDecisionTables. . . . . . 29 1. 3 PartialDecision Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

This monograph is devoted to theoretical and experimental study of partial reductsandpartialdecisionrulesonthebasisofthestudyofpartialcovers. The use of partial (approximate) reducts and decision rules instead of exact ones allowsustoobtainmorecompactdescriptionofknowledgecontainedindecision tables, andtodesignmorepreciseclassi?ers. Weconsideralgorithmsforconstructionofpartialreductsandpartialdecision rules, boundsonminimalcomplexityofpartialreductsanddecisionrules, and algorithms for construction of the set of all partial reducts and the set of all irreducible partial decision rules. We discuss results of numerous experiments with randomly generated and real-life decision tables. These results show that partial reducts and decision rules can be used in data mining and knowledge discoverybothforknowledgerepresentationandforprediction. Theresultsobtainedinthe monographcanbe usefulforresearchersinsuch areasasmachinelearning, dataminingandknowledgediscovery, especiallyfor thosewhoareworkinginroughsettheory, testtheoryandLAD(LogicalAnalysis ofData). The monographcan be usedunder the creationofcoursesforgraduates- dentsandforPh. D. studies. An essential part of software used in experiments will be accessible soon in RSES-RoughSetExplorationSystem(InstituteofMathematics, WarsawU- versity, headofproject-ProfessorAndrzejSkowron). We are greatly indebted to Professor Andrzej Skowron for stimulated d- cussionsand varioussupportof ourwork. We aregratefulto ProfessorJanusz Kacprzykforhelpfulsuggestions. Sosnowiec, Poland MikhailJu. Moshkov April2008 MarcinPiliszczuk BeataZielosko Contents Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 PartialCovers, ReductsandDecisionRules . . . . . . . . . . . . . . . . 7 1. 1 PartialCovers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1. 1. 1 MainNotions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1. 1. 2 Known Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1. 1. 3 PolynomialApproximateAlgorithms. . . . . . . . . . . . . . . . . . 10 1. 1. 4 Bounds on C (?)Based on Information about min GreedyAlgorithm Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1. 1. 5 UpperBoundon C (?). . . . . . . . . . . . . . . . . . . . . . . . . . 17 greedy 1. 1. 6 Covers fortheMostPartofSetCoverProblems. . . . . . . . 18 1. 2 PartialTests and Reducts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 1. 2. 1 MainNotions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 1. 2. 2Relationships betweenPartialCovers and Partial Tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1. 2. 3 PrecisionofGreedyAlgorithm. . . . . . . . . . . . . . . . . . . . . . . 24 1. 2. 4 PolynomialApproximateAlgorithms. . . . . . . . . . . . . . . . . . 25 1. 2. 5 Bounds on R (?)Based on Information about min GreedyAlgorithm Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 1. 2. 6 UpperBoundon R (?). . . . . . . . . . . . . . . . . . . . . . . . . . 28 greedy 1. 2. 7 Tests fortheMostPartofBinaryDecisionTables. . . . . . 29 1. 3 PartialDecision Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

This two-volume set LNAI 10313 and LNAI 10314 constitutes the proceedings of the International Joint Conference on Rough Sets, IJCRS 2017, held in Olsztyn, Poland, in July 2017. The 74 revised full papers presented together with 16 short papers and 16 invited talks, were carefully reviewed and selected from 130 submissions. The papers in this two set-volume of IJCRS 2017 follow the track already rutted by RSCTC and JRS conferences which aimed at unification of many facets of rough set theory from theoretical aspects of the rough set idea bordering on theory of concepts and going through algebraic structures, topological structures, logics for uncertain reasoning, decision algorithms, relations to other theories of vagueness and ambiguity, then to extensions of the rough set idea like granular structures, rough mereology, and to applications of the idea in diverse fields of applied science including hybrid methods like rough-fuzzy, neuro-rough, neuro-rough-fuzzy computing. IJCRS 2017 encompasses topics spread among four main tracks: Rough Sets and Data Science (in relation to RSCTC series organized since 1998); Rough Sets and Granular Computing (in relation to RSFDGrC series organized since 1999); Rough Sets and Knowledge Technology (in relation to RSKT series organized since 2006); and Rough Sets and Intelligent Systems (in relation to RSEISP series organized since 2007).

This two-volume set LNAI 10313 and LNAI 10314 constitutes the proceedings of the International Joint Conference on Rough Sets, IJCRS 2017, held in Olsztyn, Poland, in July 2017. The 74 revised full papers presented together with 16 short papers and 16 invited talks, were carefully reviewed and selected from 130 submissions. The papers in this two set-volume of IJCRS 2017 follow the track already rutted by RSCTC and JRS conferences which aimed at unification of many facets of rough set theory from theoretical aspects of the rough set idea bordering on theory of concepts and going through algebraic structures, topological structures, logics for uncertain reasoning, decision algorithms, relations to other theories of vagueness and ambiguity, then to extensions of the rough set idea like granular structures, rough mereology, and to applications of the idea in diverse fields of applied science including hybrid methods like rough-fuzzy, neuro-rough, neuro-rough-fuzzy computing. IJCRS 2017 encompasses topics spread among four main tracks: Rough Sets and Data Science (in relation to RSCTC series organized since 1998); Rough Sets and Granular Computing (in relation to RSFDGrC series organized since 1999); Rough Sets and Knowledge Technology (in relation to RSKT series organized since 2006); and Rough Sets and Intelligent Systems (in relation to RSEISP series organized since 2007).