]> BibSonomy publications for /author/Shum http://www.bibsonomy.org/burst/author/Shum BibSonomy BuRST Feed for /author/Shum 2008-07-27T02:53:19+02:00 Variational Specular Separation Using Color and Polarization. dblphttp://www.bibsonomy.org/bibtex/2a4ed822854bec397d809e61e315e11f6/dblp dblp 2008-06-30T00:00:00+02:00 dblp
Variational Specular Separation Using Color and Polarization.
Dae Woong Kim and Stephen Lin and Ki-Sang Hong and Heung-Yeung Shum MVA 176-179 (2002)
to dblp by dblp on 2008-06-30 00:00:00 ]]> MVAconf/mva/2002176-179Variational Specular Separation Using Color and Polarization.2002dblp 2008-06-30 00:00:00.0 Markov and Artin Normal Form Theorem for Braid Groups. dblphttp://www.bibsonomy.org/bibtex/2b7abe819660a21a2d8c213a6e83d60ed/dblp dblp 2008-06-23T00:00:00+02:00 dblp
Markov and Artin Normal Form Theorem for Braid Groups.
L. A. Bokut and V. V. Chaynikov and K. P. Shum IJAC 17 951-961 (2007)
to dblp by dblp on 2008-06-23 00:00:00 ]]> IJAC5/6951-961Markov and Artin Normal Form Theorem for Braid Groups.172007dblp 2008-06-23 00:00:00.0 Learning acyclic decision trees with Functional Dependency Network and MDL Genetic Programming http://www.bibsonomy.org/bibtex/2464049d61e708ccb05728e672ae7eca0/brazovayeye brazovayeye 2008-06-19T17:46:40+02:00 algorithms, genetic programming
Learning acyclic decision trees with Functional Dependency Network and MDL Genetic Programming
Wing-Ho Shum and Kwong-Sak Leung and Man-Leung Wong International Multi-Conference on Computing in the Global Information Technology, ICCGI '06 25 (2006)
to algorithms, genetic programming by brazovayeye on 2008-06-19 17:46:40 ]]> BucharestInternational Multi-Conference on Computing in the Global Information Technology, ICCGI '061-3 August25Learning acyclic decision trees with Functional Dependency Network and {MDL} Genetic Programming2006algorithms, genetic programming 2008-06-19 17:46:40.0One objective of data mining is to discover parent-child relationships among a set of variables in the domain. Moreover, showing parents' importance can further help to improve decision makings' quality. Bayesian network (BN) is a useful model for multi-class problems and can illustrate parent-child relationships with no cycle. But it cannot show parents' importance. In contrast, decision trees state parents' importance clearly, for instance, the most important parent is put in the first level. However, decision trees are proposed for single-class problems only, when they are applied to multi-class ones, they are likely to produce cycles representing tautologic. In this paper, we propose to use MDL genetic programming (MDLGP) and functional dependency network (FDN) to learn a set of acyclic decision trees (Shum et al., 2005). The FDN is an extension of BN; it can handle all of discrete, continuous, interval and ordinal values; it guarantees to produce decision trees with no cycle; its learning search space is smaller than decision trees'; and it can represent higher-order relationships among variables. The MDLGP is a robust genetic programming (GP) proposed to learn the FDN. We also propose a method to derive acyclic decision trees from the FDN. The experimental results demonstrate that the proposed method can successfully discover the target decision trees, which have no cycle and have the accurate classification results Learning non-overlapping rules A method based on Functional Dependency Network and MDL Genetic Programming http://www.bibsonomy.org/bibtex/27c7679f0ef4258297365cb6de5d0ca26/brazovayeye brazovayeye 2008-06-19T17:46:40+02:00 algorithms, genetic poster programming,
Learning non-overlapping rules A method based on Functional Dependency Network and MDL Genetic Programming
Wing-Ho Shum and Kwong-Sak Leung and Man-Leung Wong Proceedings of the 2006 IEEE Congress on Evolutionary Computation 2717--2724 (2006)
to algorithms, genetic poster programming, by brazovayeye on 2008-06-19 17:46:40 ]]> VancouverProceedings of the 2006 IEEE Congress on Evolutionary Computation6-21 July2717--2724Learning non-overlapping rules {A} method based on Functional Dependency Network and {MDL} Genetic Programming2006algorithms, genetic poster programming, 2008-06-19 17:46:40.0Classification rule is a useful model in data mining. Given variable values, rules classify data items into different classes. Different rule learning algorithms are proposed, like Genetic Algorithm (GA) and Genetic Programming (GP). Rules can also be extracted from Bayesian Network (BN) and decision trees. However, all of them have disadvantages and may fail to get the best results. Both of GA and GP cannot handle cooperation among rules and thus, the learnt rules are likely to have many overlappings, i.e. more than one rules classify the same data items and different rules have different predictions. The conflicts among the rules reduce their understandability and increase their usage difficulty for expert systems. In contrast, rules extracted from BN and decision trees have no overlapping in nature. But BN can handle discrete values only and cannot represent higher-order relationships among variables. Moreover, the search space for decision tree learning is huge and thus, it is difficult to reach the global optimum. In this paper, we propose to use Functional Dependency Network (FDN) and MDL Genetic Programming (MDLGP) to learn a set of non-overlapping classification rules [17]. The FDN is an extension of BN; it can handle all kind of values; it can represent higher-order relationships among variables; and its learning search space is smaller than decision trees'. The experimental results demonstrate that the proposed method can successfully discover the target rules, which have no overlapping and have the highest classification accuracies. Learning Functional Dependency Networks Based on Genetic Programming http://www.bibsonomy.org/bibtex/22461bee242feaaf143fb7b19fdff10ad/brazovayeye brazovayeye 2008-06-19T17:46:40+02:00 algorithms, genetic programming
Learning Functional Dependency Networks Based on Genetic Programming
Wing-Ho Shum and Kwong-Sak Leung and Man Leung Wong Proceedings of the 5th IEEE International Conference on Data Mining (ICDM 2005) 394--401 (2005)
to algorithms, genetic programming by brazovayeye on 2008-06-19 17:46:40 ]]> Houston, Texas, USAProceedings of the 5th IEEE International Conference on Data Mining (ICDM 2005)27-30 November394--401Learning Functional Dependency Networks Based on Genetic Programming2005algorithms, genetic programming 2008-06-19 17:46:40.0Bayesian Network (BN) is a powerful network model, which represents a set of variables in the domain and provides the probabilistic relationships among them. But BN can handle discrete values only; it cannot handle continuous, interval and ordinal ones, which must be converted to discrete values and the order information is lost. Thus, BN tends to have higher network complexity and lower understandability. In this paper, we present a novel dependency network which can handle discrete, continuous, interval and ordinal values through functions; it has lower network complexity and stronger expressive power; it can represent any kind of relationships; and it can incorporate a-priori knowledge though user-defined functions. We also propose a novel Genetic Programming (GP) to learn dependency networks. The novel GP does not use any knowledge-guided nor application-oriented operator, thus it is robust and easy to replicate. The experimental results demonstrate that the novel GP can successfully discover the target novel dependency networks, which have the highest accuracy and the lowest network complexity. Co-evolutionary Rule-Chaining Genetic Programming http://www.bibsonomy.org/bibtex/2abd1f5d545a13a95e02706db78392283/brazovayeye brazovayeye 2008-06-19T17:46:40+02:00 Agents Complex Systems algorithms, and genetic programming,
Co-evolutionary Rule-Chaining Genetic Programming
Wing-Ho Shum and Kwong-Sak Leung and Man Leung Wong Intelligent Data Engineering and Automated Learning - IDEAL 2005, 6th International Conference, Proceedings 546--554 (2005)
to Agents Complex Systems algorithms, and genetic programming, by brazovayeye on 2008-06-19 17:46:40 ]]> Brisbane, AustraliaIntelligent Data Engineering and Automated Learning - IDEAL 2005, 6th International Conference, ProceedingsJuly 6-8546--554Lecture Notes in Computer ScienceCo-evolutionary Rule-Chaining Genetic Programming35782005Agents Complex Systems algorithms, and genetic programming, 2008-06-19 17:46:40.0Genetic Programming (GP) paradigm called Co-evolutionary Rule-Chaining Genetic Programming (CRGP) has been proposed to learn the relationships among attributes represented by a set of classification rules for multi-class problems. It employs backward chaining inference to carry out classification based on the acquired acyclic rule set. Its main advantages are: 1) it can handle more than one class at a time; 2) it avoids cyclic result; 3) unlike Bayesian Network (BN), the CRGP can handle input attributes with continuous values directly; and 4) with the flexibility of GP, CRGP can learn complex relationship. We have demonstrated its better performance on one synthetic and one real-life medical data sets. Cohere: Towards Web 2.0 Argumentation. dblphttp://www.bibsonomy.org/bibtex/26edbebc95381d17f5250214c377624bd/dblp dblp 2008-06-15T00:00:00+02:00 dblp
Cohere: Towards Web 2.0 Argumentation.
Simon Buckingham Shum COMMA 97-108 (2008)
to dblp by dblp on 2008-06-15 00:00:00 ]]> COMMAconf/comma/200897-108Frontiers in Artificial Intelligence and ApplicationsCohere: Towards Web 2.0 Argumentation.1722008dblp 2008-06-15 00:00:00.0 Ontological Foundations for Scholarly Debate Mapping Technology. dblphttp://www.bibsonomy.org/bibtex/2bfe0146fc4708320aa742f09af1350d1/dblp dblp 2008-06-15T00:00:00+02:00 dblp
Ontological Foundations for Scholarly Debate Mapping Technology.
Neil Benn and Simon Buckingham Shum and John Domingue and Clara Mancini COMMA 61-72 (2008)
to dblp by dblp on 2008-06-15 00:00:00 ]]> COMMAconf/comma/200861-72Frontiers in Artificial Intelligence and ApplicationsOntological Foundations for Scholarly Debate Mapping Technology.1722008dblp 2008-06-15 00:00:00.0 Image-Based Motion Compensation for Structured Light Scanning of Dynamic Surfaces. dblphttp://www.bibsonomy.org/bibtex/2c430cb2757280266782c51632961d900/dblp dblp 2008-06-12T00:00:00+02:00 dblp
Image-Based Motion Compensation for Structured Light Scanning of Dynamic Surfaces.
Stefan Gumhold and Sören König Visual Computing - Convergence of Computer Graphics and Computer Vision (2007)
to dblp by dblp on 2008-06-12 00:00:00 ]]> Visual Computing - Convergence of Computer Graphics and Computer Visionconf/dagstuhl/2007P7171Dagstuhl Seminar ProceedingsImage-Based Motion Compensation for Structured Light Scanning of Dynamic Surfaces.071712007dblp 2008-06-12 00:00:00.0 07171 Summary -- Visual Computing -- Convergence of Computer Graphics and Computer Vision. dblphttp://www.bibsonomy.org/bibtex/2492fc3b5d84a1e2c7b8177fcb5e15eca/dblp dblp 2008-06-12T00:00:00+02:00 dblp
07171 Summary -- Visual Computing -- Convergence of Computer Graphics and Computer Vision.
Markus H. Gross and Heinrich Müller and Hans-Peter Seidel and Harry Shum Visual Computing - Convergence of Computer Graphics and Computer Vision (2007)
to dblp by dblp on 2008-06-12 00:00:00 ]]> Visual Computing - Convergence of Computer Graphics and Computer Visionconf/dagstuhl/2007P7171Dagstuhl Seminar Proceedings07171 Summary -- Visual Computing -- Convergence of Computer Graphics and Computer Vision.071712007dblp 2008-06-12 00:00:00.0