Aggregation of Fuzzy Conformances
Retrieving the most suitable items and sorting them downwards from the best face many challenges. The conformance measures are able to efficiently calculate similarities between the desired value and values of considered items’ attribute regardless of different data types. These measures should be suitably aggregated, because the users usually provide different preferences among atomic conformances and therefore various aggregation functions should be considered. In this paper, we examine conjunctive functions (including non t-norms) as well as averaging and hybrid ones. In the hybrid aggregation, uninorms and ordinal sums of conjunctive and disjunctive functions have shown their perspectives in aggregating conformance measures. Diverse tasks require functions of desired behaviour and properly assigned weights or parameters. Thus, the perspectives for merging aggregation functions with the machine learning to the mutual benefits are outlined.
KeywordsConformance measure Conjunctive and disjunctive functions Averaging functions Hybrid functions Data and function fitting
This paper was partially supported by the project: VEGA No. 1/0373/18 entitled “Big data analytics as a tool for increasing the competitiveness of enterprises and supporting informed decisions” supported by the Ministry of Education, Science, Research and Sport of the Slovak Republic.
- 2.Beliakov, G., Pradera, A., Calvo Sánchez, T.: Aggregation Functions: A Guide for Practitioners. Springer, Heidelberg (2007)Google Scholar
- 6.De Baets, B., Mesiar, R.: Ordinal sums of aggregation operators. In: 8th International Conference on Information Processing and Management of Uncertainty, IPMU 2000, Madrid (2000)Google Scholar
- 9.Goebel, R., Chander, A., Holzinger, K., Lecue, F., Akata, Z., Stumpf, S., Kieseberg, P., Holzinger, A.: Explainable AI: the new 42? In: Holzinger, A., Kieseberg, P., Tjoa, A., Weippl, E. (eds.) Machine Learning and Knowledge Extraction. Lecture Notes in Computer Science, LNCS, vol. 11015, pp. 295–303. Springer, Cham (2018)Google Scholar
- 17.Singh, S., Ribeiro, M., Guestrin, C.: Programs as black–box explanations. In: Workshop on Interpretable Machine Learning in Complex Systems, NIPS 2016, Barcelona (2016)Google Scholar
- 18.Snasel, V., Kromer, P., Musilek, P., Nyongesa, H.O., Husek, D.: Fuzzy modeling of user needs for improvement of web search queries. In: Annual Meeting of the North American Fuzzy Information Processing Society, NAFIPS 2007, San Diego (2007)Google Scholar
- 20.Tung, A.K.H., Zhang, R., Koudas, N., Ooi, B.C.: Similarity search: a matching based approach. In: 32nd International Conference on Very Large Data Bases, Seoul (2006)Google Scholar
- 21.Vuc̆etić, M., Hudec, M.: A flexibile approach to matching user preferences with records in datasets based on the conformance measure and aggregation functions. In: 10th International Joint Conference on Computational Intelligence, IJCCI 2018, Seville (2018)Google Scholar
- 23.Vuc̆etić, M., Vujošević, M.: A literature overview of functional dependencies in fuzzy relational database models. Technics Technol. Educ. Manag. 7, 1593–1604 (2012)Google Scholar