Mining association rules between sets of items in large databases
SIGMOD '93 Proceedings of the 1993 ACM SIGMOD international conference on Management of data
Understanding Similarity: A Joint Project for Psychology,Case-Based Reasoning, and Law
Artificial Intelligence Review
ICCBR '01 Proceedings of the 4th International Conference on Case-Based Reasoning: Case-Based Reasoning Research and Development
Semi-supervised clustering: probabilistic models, algorithms and experiments
Semi-supervised clustering: probabilistic models, algorithms and experiments
On Efficient Handling of Continuous Attributes in Large Data Bases
Fundamenta Informaticae
Automatic planning of treatment of infants with respiratory failure through rough set modeling
RSCTC'06 Proceedings of the 5th international conference on Rough Sets and Current Trends in Computing
Ontological framework for approximation
RSFDGrC'05 Proceedings of the 10th international conference on Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing - Volume Part I
Ensembles of bireducts: towards robust classification and simple representation
FGIT'11 Proceedings of the Third international conference on Future Generation Information Technology
Dynamic rule-based similarity model for DNA microarray data
Transactions on Rough Sets XV
| Hi-index | 0.00 |
This paper presents a methodology of constructing robust classifiers based on a concept called a Hierarchic Similarity Model (HSM). The hierarchic similarity is interpreted as a relation between pairs of complex objects. This relation can be derived from an information system by examining the domain related aspects of similarity. In the paper, global similarity is decomposed into many local similarities by analogy with the process of perceiving similar objects. For the purpose of estimating local relations some well-known rough sets methods are used, as well as context knowledge provided by a domain expert. Then the rules modeling interactions between local similarities are constructed and used to assess the degree of a global similarity of complex objects. The obtained relation can be used to construct classifiers which may successfully compete with other popular methods like boosted decision trees or k-NN algorithm. An implementation of the proposed models in the R script language is provided together with an empirical evaluation of the similarity based classification accuracy for some common datasets. This paper is a continuation of the research started in [1].