Applications Type-2 Membership Functions in Fuzzy Logic Systems Under Conditions of Uncertainty Input Data
DOI:
https://doi.org/10.31649/mccs2022.02Keywords:
fuzzy logic system, type-2 membership function, uncertainty input data, excpert knowledgeAbstract
This work proposes fuzzy logics system with use of math apparatus of type-2 interval fuzzy sets. The present paper formulates the task of developing a formal approach, which would enable analyzing fuzzy systems in terms of their capability to describe uncertainties of input information using interval membership functions. In contrast with the type-1 fuzzy sets-based systems, which result in membership value in the form of a single digit, the type-2 fuzzy sets allow to obtain an interval of values of the final linguistic variable. This interval is the result of the uncertainties, which are present in the incoming data, as well as of the uncertainties, related to the nature of presentation of expertise. These peculiarities allow the interval fuzzy sets-based system to function operating with less informative incoming data as well as in case of data omission. These conditions make function of the type-1 fuzzy logical systems impossible. That is why use of the fuzzy logics systems with interval functions is seen expedient compared to the usual type-1 systems. However, such models may result in wide interval output. In case the resulting output of the model may not be used to solve a task or is not satisfactory for the developer, it is better to get to use of the three-dimension membership functions while obtaining the outcome as a linguistic interpretation. This work also describes the stages of building fuzzy knowledge base using the common rule of use of experimental data. This work also proposes an approach to evaluate the parameters of gauss type-2 membership interval functions by defining the membership functions permissible measuring ranges which allow for adequate modelling. As long as use of the interval membership function provides for the fuzzy system output value with certain interval, taking into account certain vector of input values, this work provides expert recommendations on how to evaluate the interval output.
References
Zaychenko Y.P. Fuzzy models and methods in intelligent systems. Kiev Slovo (2008). (in Russian)
Liang, Q. Interval Type-2 fuzzy logic systems: theory and design [Text] / Q. Liang, J. M. Mendel // IEEE Trans. on Fuzzy Syst. – 2000. –V. 8. – P. 535–550.
Maribel Guerrero, Fevrier Valdez and Oscar Castillo. Comparative Study between Type-1 and Interval Type-2 FuzzySystems in Parameter Adaptation for the Cuckoo Search Algorithm [Text] / Maribel Guerrero, Fevrier Valdez // Symmetry 2022, 14, 2289, https://doi.org/10.3390/sym14112289.
Kondratenko, N. R. Investigating adequacy of interval type-2 fuzzy models in comlex objects identification problems [Text] / N. R. Kondratenko, O. O. Snihur // System Research And Information Technologies. – 2019. – NO 4. – P. 94–104.
Kondratenko N.R. Vykorystannya nechitkykh baz znanj z funktsiyamy nalezhnosti typu-2 u medychniy diahnostytsi / N.R. Kondratenko // Materialy mizhnar. nauk.-prakt. konf., «Аktual’ni zadachi medychnoyi biolohichnoyi fizyky ta informatyky»: - 2022р.:.- Vinnytsya, rezhym dostupu: : https://meet.google/com/pap-cszd-zpi
Kondratenko, N. R. Fuzzy Logic Systems with Allovance for the Blank in Experimental Data Taken [Text] / N. R. Kondratenko, N. B. Zelinsjka, S.M. Kuzemko // Naukovi visti NTUU KPI.- 2004.- No. 5. P. 37–41.
Kondratenko, N. R. Improving Adequacy of Type-2 Fuzzy Models by Using Type-2 Fuzzy Sets [Text] / N. Kondratenko // Naukovi visti NTUU KPI. – 2014. – No. 6. – P. 56–61.
Kondratenko, N. Interval Fuzzy Modeling of Comlex Systems under Conditions of Input Data Uncertainly [Text] / N. Kondratenko, О. Snihur // Eastern-European Journal of Enterprise Technologies. – 2016. – V. 4/4 (82). – P. 20–28.
