Two-way Parallel Class Expression Learning

An C. Tran, Jens Dietrich, Hans W. Guesgen, Stephen Marsland
Proceedings of the Asian Conference on Machine Learning, PMLR 25:443-458, 2012.

Abstract

In machine learning, we often encounter datasets that can be described using simple rules and regular exception patterns describing situations where those rules do not apply. In this paper, we propose a two-way parallel class expression learning algorithm that is suitable for this kind of problem. This is a top-down refinement-based class expression learning algorithm for Description Logic (DL). It is distinguished from similar DL learning algorithms in the way it uses the concepts generated by the refinement operator. In our approach, we unify the computation of concepts describing positive and negative examples, but we maintain them separately, and combine them at the end. By doing so, we can avoid the use of negation in the refinement without any loss of generality. Evaluation shows that our approach can reduce the search space significantly, and therefore the learning time is reduced. Our implementation is based on the DL-Learner framework and we inherit the Parallel Class Expression Learning (ParCEL) algorithm design for parallelisation.

Cite this Paper

BibTeX
@InProceedings{pmlr-v25-tran12c, title = {Two-way Parallel Class Expression Learning}, author = {Tran, An C. and Dietrich, Jens and Guesgen, Hans W. and Marsland, Stephen}, booktitle = {Proceedings of the Asian Conference on Machine Learning}, pages = {443--458}, year = {2012}, editor = {Hoi, Steven C. H. and Buntine, Wray}, volume = {25}, series = {Proceedings of Machine Learning Research}, address = {Singapore Management University, Singapore}, month = {04--06 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v25/tran12c/tran12c.pdf}, url = {https://proceedings.mlr.press/v25/tran12c.html}, abstract = {In machine learning, we often encounter datasets that can be described using simple rules and regular exception patterns describing situations where those rules do not apply. In this paper, we propose a two-way parallel class expression learning algorithm that is suitable for this kind of problem. This is a top-down refinement-based class expression learning algorithm for Description Logic (DL). It is distinguished from similar DL learning algorithms in the way it uses the concepts generated by the refinement operator. In our approach, we unify the computation of concepts describing positive and negative examples, but we maintain them separately, and combine them at the end. By doing so, we can avoid the use of negation in the refinement without any loss of generality. Evaluation shows that our approach can reduce the search space significantly, and therefore the learning time is reduced. Our implementation is based on the DL-Learner framework and we inherit the Parallel Class Expression Learning (ParCEL) algorithm design for parallelisation.} }
Endnote
%0 Conference Paper %T Two-way Parallel Class Expression Learning %A An C. Tran %A Jens Dietrich %A Hans W. Guesgen %A Stephen Marsland %B Proceedings of the Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2012 %E Steven C. H. Hoi %E Wray Buntine %F pmlr-v25-tran12c %I PMLR %P 443--458 %U https://proceedings.mlr.press/v25/tran12c.html %V 25 %X In machine learning, we often encounter datasets that can be described using simple rules and regular exception patterns describing situations where those rules do not apply. In this paper, we propose a two-way parallel class expression learning algorithm that is suitable for this kind of problem. This is a top-down refinement-based class expression learning algorithm for Description Logic (DL). It is distinguished from similar DL learning algorithms in the way it uses the concepts generated by the refinement operator. In our approach, we unify the computation of concepts describing positive and negative examples, but we maintain them separately, and combine them at the end. By doing so, we can avoid the use of negation in the refinement without any loss of generality. Evaluation shows that our approach can reduce the search space significantly, and therefore the learning time is reduced. Our implementation is based on the DL-Learner framework and we inherit the Parallel Class Expression Learning (ParCEL) algorithm design for parallelisation.
RIS
TY - CPAPER TI - Two-way Parallel Class Expression Learning AU - An C. Tran AU - Jens Dietrich AU - Hans W. Guesgen AU - Stephen Marsland BT - Proceedings of the Asian Conference on Machine Learning DA - 2012/11/17 ED - Steven C. H. Hoi ED - Wray Buntine ID - pmlr-v25-tran12c PB - PMLR DP - Proceedings of Machine Learning Research VL - 25 SP - 443 EP - 458 L1 - http://proceedings.mlr.press/v25/tran12c/tran12c.pdf UR - https://proceedings.mlr.press/v25/tran12c.html AB - In machine learning, we often encounter datasets that can be described using simple rules and regular exception patterns describing situations where those rules do not apply. In this paper, we propose a two-way parallel class expression learning algorithm that is suitable for this kind of problem. This is a top-down refinement-based class expression learning algorithm for Description Logic (DL). It is distinguished from similar DL learning algorithms in the way it uses the concepts generated by the refinement operator. In our approach, we unify the computation of concepts describing positive and negative examples, but we maintain them separately, and combine them at the end. By doing so, we can avoid the use of negation in the refinement without any loss of generality. Evaluation shows that our approach can reduce the search space significantly, and therefore the learning time is reduced. Our implementation is based on the DL-Learner framework and we inherit the Parallel Class Expression Learning (ParCEL) algorithm design for parallelisation. ER -
APA
Tran, A.C., Dietrich, J., Guesgen, H.W. & Marsland, S.. (2012). Two-way Parallel Class Expression Learning. Proceedings of the Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 25:443-458 Available from https://proceedings.mlr.press/v25/tran12c.html.

Related Material