scispace - formally typeset
Search or ask a question
Proceedings Article

Machine Learning by Function Decomposition

TL;DR: A new machine learning method is presented that, given a set of training examples, induces a definition of the target concept in terms of a hierarchy of intermediate concepts and their definitions, which effectively decomposes the problem into smaller, less complex problems.
Abstract: We present a new machine learning method that, given a set of training examples, induces a definition of the target concept in terms of a hierarchy of intermediate concepts and their definitions. This effectively decomposes the problem into smaller, less complex problems. The method is inspired by the Boolean function decomposition approach to the design of digital circuits. To cope with high time complexity of finding an optimal decomposition, we propose a suboptimal heuristic algorithm. The method, implemented in program HINT (HIerarchy Induction Tool), is experimentally evaluated using a set of artificial and real-world learning problems. It is shown that the method performs well both in terms of classification accuracy and discovery of meaningful concept hierarchies.
Citations
More filters
Journal ArticleDOI
TL;DR: Results show that the data encoded with Sum Coding and Backward Difference Coding technique give highest accuracy as compared to the data pre-processed by rest of the techniques.
Abstract: In classification analysis, the dependent variable is frequently influenced not only by ratio scale variables, but also by qualitative (nominal scale) variables. Machine Learning algorithms accept only numerical inputs, hence, it is necessary to encode these categorical variables into numerical values using encoding techniques. This paper presents a comparative study of seven categorical variable encoding techniques to be used for classification using Artificial Neural Networks on a categorical dataset. The Car Evaluation dataset provided by UCI is used for training. Results show that the data encoded with Sum Coding and Backward Difference Coding technique give highest accuracy as compared to the data pre-processed by rest of the techniques.

332 citations

01 Jun 1961
TL;DR: In this article, the Ashenhurst chart method is generalized to non-junctive decompositions by means of the don't care conditions, which leads to designs of more economical switching circuits to realize the given switching function.
Abstract: : A given switching function of n variables can frequently be decomposed into a composite function of several essentially simpler switching functions. Such decompositions lead to designs of more economical switching circuits to realize the given switching function. Ashenhurst's chart method is generalized to nondisjunctive decompositions by means of the don't care conditions. This extension provides an effective method of constructing all decompositions of switching functions. (Author)

227 citations

01 Apr 2004
TL;DR: This thesis is focused on the monotonicity property in knowledge discovery and more specifically in classification, attribute reduction, function decomposition, frequent patterns generation and missing values handling.
Abstract: textThe monotonicity property is ubiquitous in our lives and it appears in different roles: as domain knowledge, as a requirement, as a property that reduces the complexity of the problem, and so on. It is present in various domains: economics, mathematics, languages, operations research and many others. This thesis is focused on the monotonicity property in knowledge discovery and more specifically in classification, attribute reduction, function decomposition, frequent patterns generation and missing values handling. Four specific problems are addressed within four different methodologies, namely, rough sets theory, monotone decision trees, function decomposition and frequent patterns generation. In the first three parts, the monotonicity is domain knowledge and a requirement for the outcome of the classification process. The three methodologies are extended for dealing with monotone data in order to be able to guarantee that the outcome will also satisfy the monotonicity requirement. In the last part, monotonicity is a property that helps reduce the computation of the process of frequent patterns generation. Here the focus is on two of the best algorithms and their comparison both theoretically and experimentally. About the Author: Viara Popova was born in Bourgas, Bulgaria in 1972. She followed her secondary education at Mathematics High School "Nikola Obreshkov" in Bourgas. In 1996 she finished her higher education at Sofia University, Faculty of Mathematics and Informatics where she graduated with major in Informatics and specialization in Information Technologies in Education. She then joined the Department of Information Technologies, First as an associated member and from 1997 as an assistant professor. In 1999 she became a PhD student at Erasmus University Rotterdam, Faculty of Economics, Department of Computer Science. In 2004 she joined the Artificial Intelligence Group within the Department of Computer Science, Faculty of Sciences at Vrije Universiteit Amsterdam as a PostDoc researcher.

103 citations

Book
23 Nov 2004
TL;DR: This book discusses the theory and application of Ontology Translation Approaches for Interoperability, and the development of Semantic Web Services and Problem Solving Methods.
Abstract: Ontologies: Mappings and Translation.- The Theory of Top-Level Ontological Mappings and Its Application to Clinical Trial Protocols.- Generating and Integrating Evidence for Ontology Mappings.- Ontology Translation Approaches for Interoperability: A Case Study with Protege-2000 and WebODE.- Ontologies: Problems and Applications.- On the Foundations of UML as an Ontology Representation Language.- OWL Pizzas: Practical Experience of Teaching OWL-DL: Common Errors & Common Patterns.- Using a Novel ORM-Based Ontology Modelling Method to Build an Experimental Innovation Router.- Ontology-Based Functional-Knowledge Modeling Methodology and Its Deployment.- Ontologies: Trust and E-learning.- Accuracy of Metrics for Inferring Trust and Reputation in Semantic Web-Based Social Networks.- Semantic Webs for Learning: A Vision and Its Realization.- Ontology Maintenance.- Enhancing Ontological Knowledge Through Ontology Population and Enrichment.- Refactoring Methods for Knowledge Bases.- Applications to Medicine.- Managing Patient Record Instances Using DL-Enabled Formal Concept Analysis.- Medical Ontology and Virtual Staff for a Health Network.- Portals.- A Semantic Portal for the International Affairs Sector.- OntoWeaver-S: Supporting the Design of Knowledge Portals.- Knowledge Acquisition.- Graph-Based Acquisition of Expressive Knowledge.- Incremental Knowledge Acquisition for Improving Probabilistic Search Algorithms.- Parallel Knowledge Base Development by Subject Matter Experts.- Designing a Procedure for the Acquisition of Probability Constraints for Bayesian Networks.- Invented Predicates to Reduce Knowledge Acquisition.- Web Services and Problem Solving Methods.- Extending Semantic-Based Matchmaking via Concept Abduction and Contraction.- Configuration of Web Services as Parametric Design.- Knowledge Modelling for Deductive Web Mining.- On the Knowledge Level of an On-line Shop Assistant.- A Customer Notification Agent for Financial Overdrawn Using Semantic Web Services.- Aggregating Web Services with Active Invocation and Ensembles of String Distance Metrics.- Search, Browsing and Knowledge Acquisition.- KATS: A Knowledge Acquisition Tool Based on Electronic Document Processing.- SERSE: Searching for Digital Content in Esperonto.- A Topic-Based Browser for Large Online Resources.- Knowledge Formulation for AI Planning.- Short Papers.- ConEditor: Tool to Input and Maintain Constraints.- Adaptive Link Services for the Semantic Web.- Using Case-Based Reasoning to Support Operational Knowledge Management.- A Hybrid Algorithm for Alignment of Concept Hierarchies.- Cultural Heritage Information on the Semantic Web.- Stepper: Annotation and Interactive Stepwise Transformation for Knowledge-Rich Documents.- Knowledge Management and Interactive Learning.- Ontology-Based Semantic Annotations for Biochip Domain.- Toward a Library of Problem-Solving Methods on the Internet.- Supporting Collaboration Through Semantic-Based Workflow and Constraint Solving.- Towards a Knowledge-Aware Office Environment.- Computing Similarity Between XML Documents for XML Mining.- A CBR Driven Genetic Algorithm for Microcalcification Cluster Detection.- Ontology Enrichment Evaluation.- KAFTIE: A New KA Framework for Building Sophisticated Information Extraction Systems.- From Text to Ontology: The Modelling of Economics Events.- Discovering Conceptual Web-Knowledge in Web Documents.- Knowledge Mediation: A Procedure for the Cooperative Construction of Domain Ontologies.- A Framework to Improve Semantic Web Services Discovery and Integration in an E-Gov Knowledge Network.- Knowledge Organisation and Information Retrieval with Galois Lattices.- Acquisition of Causal and Temporal Knowledge in Medical Domains. A Web-Based Approach.

101 citations


Cites background or methods or result from "Machine Learning by Function Decomp..."

  • ..., [27]), (technological) innovation has been recognised as a crucial part of a company’s assets....

    [...]

  • ...By role we mean here such a concept that an entity plays in a specific context and cannot be defined without mentioning external concepts [26], which is similar to the definitions in the literature [27,28]....

    [...]

  • ...Some of the major works as regards appearance in the literature are FCA-Merge [26], OntoMorph [27], Chimaera [28] and the tools of the PROMPT suite [25], which is developed at Stanford University....

    [...]

Proceedings Article
26 Jul 2005
TL;DR: In this article, knowledge about qualitative monotonicities was formally represented and incorporated into learning algorithms, and they showed that using knowledge of qualitative influences as constraints on probability distributions yields improved accuracy.
Abstract: When training data is sparse, more domain knowledge must be incorporated into the learning algorithm in order to reduce the effective size of the hypothesis space. This paper builds on previous work in which knowledge about qualitative monotonicities was formally represented and incorporated into learning algorithms (e.g., Clark & Matwin's work with the CN2 rule learning algorithm). We show how to interpret knowledge of qualitative influences, and in particular of monotonicities, as constraints on probability distributions, and to incorporate this knowledge into Bayesian network learning algorithms. We show that this yields improved accuracy, particularly with very small training sets (e.g. less than 10 examples).

99 citations

References
More filters
Book
15 Oct 1992
TL;DR: A complete guide to the C4.5 system as implemented in C for the UNIX environment, which starts from simple core learning methods and shows how they can be elaborated and extended to deal with typical problems such as missing data and over hitting.
Abstract: From the Publisher: Classifier systems play a major role in machine learning and knowledge-based systems, and Ross Quinlan's work on ID3 and C4.5 is widely acknowledged to have made some of the most significant contributions to their development. This book is a complete guide to the C4.5 system as implemented in C for the UNIX environment. It contains a comprehensive guide to the system's use , the source code (about 8,800 lines), and implementation notes. The source code and sample datasets are also available on a 3.5-inch floppy diskette for a Sun workstation. C4.5 starts with large sets of cases belonging to known classes. The cases, described by any mixture of nominal and numeric properties, are scrutinized for patterns that allow the classes to be reliably discriminated. These patterns are then expressed as models, in the form of decision trees or sets of if-then rules, that can be used to classify new cases, with emphasis on making the models understandable as well as accurate. The system has been applied successfully to tasks involving tens of thousands of cases described by hundreds of properties. The book starts from simple core learning methods and shows how they can be elaborated and extended to deal with typical problems such as missing data and over hitting. Advantages and disadvantages of the C4.5 approach are discussed and illustrated with several case studies. This book and software should be of interest to developers of classification-based intelligent systems and to students in machine learning and expert systems courses.

21,674 citations


"Machine Learning by Function Decomp..." refers methods in this paper

  • ...5 inductive decision tree learner (Quinlan 1993) run on the same data....

    [...]

01 Jan 1998

12,940 citations


"Machine Learning by Function Decomp..." refers background in this paper

  • ...MONK1 and MONK2 Well-known six-attributebinary classi cation problems taken from thesame repository (Murphy and Aha 1994, Thrunet al. 1991)....

    [...]

  • ...LENSES A small domain taken from UCI machinelearning repository (Murphy and Aha 1994)....

    [...]

Journal ArticleDOI
A. L. Samuel1
TL;DR: Full use is made of the so called "alpha-beta" pruning and several forms of forward pruning to restrict the spread of the move tree and to permit the program to look ahead to a much greater depth than it otherwise could do.
Abstract: A new signature tablet echnique is described together with an improved book learning procedure which is thougthot be much superior to the linear polynomial method described earlier. Full use is made of the so called "alpha-beta" pruning and several forms of forward pruning to restrict the spread of the move tree and to permit the programt o look ahead to a much greater depth than it otherwise could do. While still unable to outplay checker masters, the program's playing ability has been greatly improved.

591 citations


"Machine Learning by Function Decomp..." refers methods in this paper

  • ...He proposed a method based on a signature table sys-tem (Samuel 1967) and successfully used it as an eval-uation mechanism for his checkers playing programs....

    [...]

01 Jun 1961
TL;DR: In this article, the Ashenhurst chart method is generalized to non-junctive decompositions by means of the don't care conditions, which leads to designs of more economical switching circuits to realize the given switching function.
Abstract: : A given switching function of n variables can frequently be decomposed into a composite function of several essentially simpler switching functions. Such decompositions lead to designs of more economical switching circuits to realize the given switching function. Ashenhurst's chart method is generalized to nondisjunctive decompositions by means of the don't care conditions. This extension provides an effective method of constructing all decompositions of switching functions. (Author)

227 citations


"Machine Learning by Function Decomp..." refers background or methods in this paper

  • ...Themethod is based on function decomposition, an ap-proach originally developed for the design of digitalcircuits (Ashenhurst 1952, Curtis 1962)....

    [...]

  • ...Ashenhurst(1952) reported on a uni ed theory of decompositionof switching functions....

    [...]

Book
01 Jan 1987
TL;DR: When you read more every page of this structured induction in expert systems, what you will obtain is something great.
Abstract: Read more and get great! That's what the book enPDFd structured induction in expert systems will give for every reader to read this book. This is an on-line book provided in this website. Even this book becomes a choice of someone to read, many in the world also loves it so much. As what we talk, when you read more every page of this structured induction in expert systems, what you will obtain is something great.

163 citations


"Machine Learning by Function Decomp..." refers background in this paper

  • ...A well-known example is struc-tured induction (a term introduced by Donald Michie) applied by Shapiro (1987)....

    [...]

  • ...In machinelearning, this principle is a foundation for structuredinduction (Shapiro 1987): instead of learning a sin-gle complex classi cation rule from examples, de ne agoal-subgoal hierarchy and learn the rules for each ofthe subgoals....

    [...]