144,622 Hits in 5.3 sec

Data-Driven Theory Refinement Using KBDistAl [chapter]

Jihoon Yang, Rajesh Parekh, Vasant Honavar, Drena Dobbs
1999 Lecture Notes in Computer Science  
We present an efficient algorithm for data-driven knowledge discovery and theory refinement using DistAl, a novel (inter-pattern distance based, polynomial time) constructive neural network learning algorithm  ...  The initial domain theory comprising of propositional rules is translated into a knowledge based network.  ...  to Drena Dobbs and Vasant Honavar.  ... 
doi:10.1007/3-540-48412-4_28 fatcat:quk3denhprdhvebgfiudu2l5x4

Refinement of uncertain rule bases via reduction

Charles X.F. Ling, Marco Valtorta
1995 International Journal of Approximate Reasoning  
However, we show that there is a class of reducible rule bases in which the strength refinement problem is NP-hard in the deep rule base, reduction is polynomial, and the fiat rule base can be refined  ...  We outline a model of rule bases with uncertainty, and give necessa~ and sufficient conditions on uncertainty combination functions that permit reduction from deep to flat (nonchaining) rule bases.  ...  ACKNOWLEDGMENTS The authors wish to thank Mike Dawes for many discussions on conditions for performing reduction, Don Loveland for many general and  ... 
doi:10.1016/0888-613x(95)00035-f fatcat:lemwrbdkgzhgzfekytgaoznvze

Combining symbolic and neural learning

Jude W. Shavlik
1994 Machine Learning  
This work was partially supported by Office of Naval Research Grants N00014-90-J-1941 and N00014-93-1-0998, National Science Foundation Grant IRI-9002413, and Department of Energy Grant DE-FG02-91ER61129  ...  Discussions with Ray Mooney, Tom Dietterich, and Geoff Hinton also substantially influenced this discussion.  ...  Towell (1992) has shown that KBANN's knowledge-based networks better refine a domain theory than do purely symbolic theory-refinement systems.  ... 
doi:10.1007/bf00993982 fatcat:qrk6g7xq7bewnizjftt3wwdnhe

A fuzzy theory refinement algorithm

Antonio Gonzalez, Raúl Perez
1998 International Journal of Approximate Reasoning  
A fuzzy theory refinement algorithm composed of a heuristic process of generalization, specification, addition and elimination of rules is proposed.  ...  This refinement algorithm can be applied to knowledge bases obtained from several sources (learning algorithms, experts), but its development is strongly associated with the SLAVE learning system.  ...  We want to know the behavior of this refinement algorithm using different theories for this problem. Therefore, we propose the following knowledge bases: UnK.  ... 
doi:10.1016/s0888-613x(98)00013-9 fatcat:qaxilp7i2zb5xm7gl7w7dlb56i

A Re-evaluation of the Over-Searching Phenomenon in Inductive Rule Learning [chapter]

Frederik Janssen, Johannes Fürnkranz
2009 Proceedings of the 2009 SIAM International Conference on Data Mining  
In this paper, we evaluate the spectrum of different search strategies to see whether separate-and-conquer rule learning algorithms are able to gain performance in terms of predictive accuracy or theory  ...  Unlike previous results that demonstrated that rule learning algorithms suffer from over-searching, our work pays particular attention to the interaction between the search heuristic and the search strategy  ...  Therefore we concentrated on evaluating different strategies for searching a single rule and combine them into a rule-based theory via a covering loop.  ... 
doi:10.1137/1.9781611972795.29 dblp:conf/sdm/JanssenF09 fatcat:6lakl6eujbchnn36hyeirij3na

Page 2378 of Mathematical Reviews Vol. , Issue 95d [page]

1995 Mathematical Reviews  
This article presents an interesting and powerful theory, named “Either”, for knowledge base refinement.  ...  {For the entire collection see MR 95c:68016.} 95d:68140 68T30 Ourston, Dirk; Mooney, Raymond J. (1-TX-C; Austin, TX) Theory refinement combining analytical and empirical methods.  ... 

Cascade ARTMAP: integrating neural computation and symbolic knowledge processing

Ah-Hwee Tan
1997 IEEE Transactions on Neural Networks  
Benchmark study on a DNA promoter recognition problem shows that with the added advantage of fast learning, cascade ARTMAP rule insertion and refinement algorithms produce performance superior to those  ...  Cascade ARTMAP, a generalization of fuzzy ARTMAP, represents intermediate attributes and rule cascades of rule-based knowledge explicitly and performs multistep inferencing.  ...  Grossberg, and M. Cohen, for their guidance and support. Thanks also go to the three anonymous referees who provided valuable comments and suggestions.  ... 
doi:10.1109/72.557661 pmid:18255628 fatcat:n5d5cirlqnd2xig6p5cycnxtzy

Theory refinement combining analytical and empirical methods

Dirk Ourston, Raymond J. Mooney
1994 Artificial Intelligence  
This article describes a comprehensive system for automatic theory (knowledge base) refinement. The system applies to classification tasks employing a propositional Hornclause domain theory.  ...  Ourston, D. and R.J. Mooney, Theory refinement combining analytical and empirical methods, Artificial Intelligence 66 (1994) 273-309.  ...  diagnosis data; Jeff Mahoney for translating the soybean theory and data and implementing the flexible tester; and Hwee Tou Ng for providing the abduction component.  ... 
doi:10.1016/0004-3702(94)90028-0 fatcat:7yvptfjpurc5vpmrgyhyrrammy

A new approach for constructing the concept map

Shian-Shyong Tseng, Pei-Chi Sue, Jun-Ming Su, Jui-Feng Weng, Wen-Nung Tsai
2007 Computers & Education  
Therefore, in Phase 1, we apply Fuzzy Set Theory to transform the numeric testing records of learners into symbolic data, apply Education Theory to further refine it, and apply Data Mining approach to  ...  Then, in Phase 2, based upon our observation in real learning situation, we use multiple rule 0360-1315/$ -see front matter Ó & Education 49 (2007) 691-707 types to further analyze the mined rules and  ...  Acknowledgement This research was partially supported by National Science Council of Republic of China under the number of NSC94-2524-S009-001, NSC94-2524-S009-002, and NSC 93-2524-S-009-004-EC3.  ... 
doi:10.1016/j.compedu.2005.11.020 fatcat:moh4cnmncjd4zexjfx32pj5ple


Johannes Fürnkranz
2012 Artificial Intelligence Review  
This paper is a survey of inductive rule learning algorithms that use a separate-andconquer strategy.  ...  We will put this wide variety of algorithms into a single framework and analyze them along three different dimensions, namely their search, language and overfitting avoidance biases.  ...  The classical separate-andconquer algorithms induce rule sets for attribute-value based concept learning problems.  ... 
doi:10.1023/a:1006524209794 fatcat:v4ax4yqgvvbd7fxzj25fayflfq

Transfer Learning via Relational Type Matching

Raksha Kumaraswamy, Phillip Odom, Kristian Kersting, David Leake, Sriraam Natarajan
2015 2015 IEEE International Conference on Data Mining  
Second, it transfers the logic rules and learns the parameters of the transferred rules using target data. Finally, it refines the rules as necessary using theory refinement.  ...  Our experimental evidence supports that this transfer method finds models as good or better than those found with state-of-the-art methods, with and without transfer, and in a fraction of the time.  ...  We thank Jan Van Haaren and Jesse Davis for sharing their TODTLER code and data.  ... 
doi:10.1109/icdm.2015.138 dblp:conf/icdm/KumaraswamyOKLN15 fatcat:hcqgbbakk5gprng2gm4jfqnyw4

Extracting refined rules from knowledge-based neural networks

Geoffrey G. Towell, Jude W. Shavlik
1993 Machine Learning  
by methods that directly refine symbolic rules; 3) are superior to those produced by previous techniques for extracting rules from trained neural networks; and 4) are "human comprehensible" Thus, this  ...  This step changes the representation of the rules from symbolic to neurally based, thereby making the rules refinable by standard neural learning methods.  ...  We wish to thank Michiel Noordewier for his construction of the two biological rule and data sets, and for general comments on this work. Comments by Richard  ... 
doi:10.1007/bf00993103 fatcat:zpid2oi6yzhavmonvzx4cvqc7q

Using Knowledge-Based Neural Networks to Improve Algorithms: Refining the Chou-Fasman Algorithm for Protein Folding [chapter]

Richard Maclin, Jude W. Shavlik
1993 Multistrategy Learning  
The extended system, FSKBANN, allows one to refine the large class of algorithms that can be represented as state-based processes.  ...  This paper describes a connectionist method for refining algorithms represented as generalized finite-state automata.  ...  The domain theory and data used in testing are available by anonymous ftp from the University of California-Irvine archive maintained by David Aha and Patrick Murphy.  ... 
doi:10.1007/978-1-4615-3202-6_5 fatcat:miehwrewzfcrlfbtd55oqhjvdy

Recommendation Using Analysis of Semantic Social Network in Social Network Services

Sangun Park, Juyoung Kang
2016 ICIC Express Letters  
This paper suggests a semantic social network that represents the various relationships between customers and products through a semantic graph, and a method that generates recommendation rules through  ...  As Social Network Services became one of the most successful Web-based businesses, recommendation using the social network also became actively utilized to attract and retain customers.  ...  The network mining algorithm is suggested based on a decision tree learning algorithm that is newly developed by defining new refinements based on description logic.  ... 
doi:10.24507/icicel.10.03.547 fatcat:uljdv5x2bndudnpypexzpnhnzm

Introduction [chapter]

Bernhard Möller, Helmut A. Partsch, Stephen A. Schuman
1993 Lecture Notes in Computer Science  
SMITH's approach to algorithm design is essentially based on the idea of translating one theory into another.  ...  BIRD/DE MOOR introduce a calculus based on a categorical setting and involving relational concepts and the theory of free inductive data types.  ...  Important concepts are problem theories, algorithm theories, program schemes as parameterized theories, design as interpretation between theories (theory morphisms), algorithm design tactics and refinement  ... 
doi:10.1007/3-540-57499-9_14 fatcat:eztaayoch5crzf76amxx6q7r3i
« Previous Showing results 1 — 15 out of 144,622 results