A kernel-based framework for learning graded relations from data
- Author
- Willem Waegeman (UGent) , Tapio Pahikkala, Antti Airola, Tapio Salakoski, Michiel Stock (UGent) and Bernard De Baets (UGent)
- Organization
- Abstract
- Driven by a large number of potential applications in areas, such as bioinformatics, information retrieval, and social network analysis, the problem setting of inferring relations between pairs of data objects has recently been investigated intensively in the machine learning community. To this end, current approaches typically consider datasets containing crisp relations so that standard classification methods can be adopted. However, relations between objects like similarities and preferences are often expressed in a graded manner in real-world applications. A general kernel-based framework for learning relations from data is introduced here. It extends existing approaches because both crisp and graded relations are considered, and it unifies existing approaches because different types of graded relations can be modeled, including symmetric and reciprocal relations. This framework establishes important links between recent developments in fuzzy set theory and machine learning. Its usefulness is demonstrated through various experiments on synthetic and real-world data. The results indicate that incorporating domain knowledge about relations improves the predictive performance.
- Keywords
- graded relations, Fuzzy relations, kernel methods, learning in graphs, machine learning, reciprocal relations, transitivity, ROCK-PAPER-SCISSORS, SUPPORT VECTOR MACHINES, PROMOTES BIODIVERSITY, RECIPROCAL RELATIONS, CYCLE-TRANSITIVITY, BACTERIAL GAME, FUZZY, PREFERENCES, NETWORK, PREDICTION
Downloads
-
KERMIT-A1-276.pdf
- full text
- |
- open access
- |
- |
- 681.12 KB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-3132065
- MLA
- Waegeman, Willem, et al. “A Kernel-Based Framework for Learning Graded Relations from Data.” IEEE TRANSACTIONS ON FUZZY SYSTEMS, vol. 20, no. 6, 2012, pp. 1090–101, doi:10.1109/TFUZZ.2012.2194151.
- APA
- Waegeman, W., Pahikkala, T., Airola, A., Salakoski, T., Stock, M., & De Baets, B. (2012). A kernel-based framework for learning graded relations from data. IEEE TRANSACTIONS ON FUZZY SYSTEMS, 20(6), 1090–1101. https://doi.org/10.1109/TFUZZ.2012.2194151
- Chicago author-date
- Waegeman, Willem, Tapio Pahikkala, Antti Airola, Tapio Salakoski, Michiel Stock, and Bernard De Baets. 2012. “A Kernel-Based Framework for Learning Graded Relations from Data.” IEEE TRANSACTIONS ON FUZZY SYSTEMS 20 (6): 1090–1101. https://doi.org/10.1109/TFUZZ.2012.2194151.
- Chicago author-date (all authors)
- Waegeman, Willem, Tapio Pahikkala, Antti Airola, Tapio Salakoski, Michiel Stock, and Bernard De Baets. 2012. “A Kernel-Based Framework for Learning Graded Relations from Data.” IEEE TRANSACTIONS ON FUZZY SYSTEMS 20 (6): 1090–1101. doi:10.1109/TFUZZ.2012.2194151.
- Vancouver
- 1.Waegeman W, Pahikkala T, Airola A, Salakoski T, Stock M, De Baets B. A kernel-based framework for learning graded relations from data. IEEE TRANSACTIONS ON FUZZY SYSTEMS. 2012;20(6):1090–101.
- IEEE
- [1]W. Waegeman, T. Pahikkala, A. Airola, T. Salakoski, M. Stock, and B. De Baets, “A kernel-based framework for learning graded relations from data,” IEEE TRANSACTIONS ON FUZZY SYSTEMS, vol. 20, no. 6, pp. 1090–1101, 2012.
@article{3132065,
abstract = {{Driven by a large number of potential applications in areas, such as bioinformatics, information retrieval, and social network analysis, the problem setting of inferring relations between pairs of data objects has recently been investigated intensively in the machine learning community. To this end, current approaches typically consider datasets containing crisp relations so that standard classification methods can be adopted. However, relations between objects like similarities and preferences are often expressed in a graded manner in real-world applications. A general kernel-based framework for learning relations from data is introduced here. It extends existing approaches because both crisp and graded relations are considered, and it unifies existing approaches because different types of graded relations can be modeled, including symmetric and reciprocal relations. This framework establishes important links between recent developments in fuzzy set theory and machine learning. Its usefulness is demonstrated through various experiments on synthetic and real-world data. The results indicate that incorporating domain knowledge about relations improves the predictive performance.}},
author = {{Waegeman, Willem and Pahikkala, Tapio and Airola, Antti and Salakoski, Tapio and Stock, Michiel and De Baets, Bernard}},
issn = {{1063-6706}},
journal = {{IEEE TRANSACTIONS ON FUZZY SYSTEMS}},
keywords = {{graded relations,Fuzzy relations,kernel methods,learning in graphs,machine learning,reciprocal relations,transitivity,ROCK-PAPER-SCISSORS,SUPPORT VECTOR MACHINES,PROMOTES BIODIVERSITY,RECIPROCAL RELATIONS,CYCLE-TRANSITIVITY,BACTERIAL GAME,FUZZY,PREFERENCES,NETWORK,PREDICTION}},
language = {{eng}},
number = {{6}},
pages = {{1090--1101}},
title = {{A kernel-based framework for learning graded relations from data}},
url = {{http://doi.org/10.1109/TFUZZ.2012.2194151}},
volume = {{20}},
year = {{2012}},
}
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: