Abstract. 

The article proposes an approach to building granular models directly on the basis of information granules expressed both in the input and output spaces. By linking these information granules, the constructed granular models are represented within neural networks in a system that includes three levels: input granules, output schema, and output granules. An extended principle of reasonable granularity is applied to construct information granules in the input space. This principle creates information granules not only by striking a reasonable balance between the two criteria, coverage and specificity, but also by optimizing these information granules based on their homogeneity, assessed with respect to the data localized in the output space. An inference scheme is assumed by analyzing the location of the input data in relation to the already formed information granules in the input space. The calculated ratio can be quantified as degrees of membership, resulting in aggregation results that include information granules in the output space. The high efficiency of the proposed granular model is ensured by the mechanisms of granular computations and the principle of justified granularity. Experimental studies have been carried out on synthetic and public data and some benchmarking has been done using rule-based models.

Keywords: 

information, neural network structures, modeling, network architecture, optimization, information granules.

PP. 81-90.

DOI: 10.14357/20790279220308

 

References

1. Yao J., Vasilakos A.V., Pedrycz W. 2013. Granular computing: Perspectives and challenges. IEEE Trans. Cybern. 43 (6):1977–1989.

2. Myung-Won Lee, Keun-Chang Kwak. 2020. Optimization by Context Refinement for Development of Incremental Granular Models. Symmetry. 12:1916.

3. Zhu X., Pedrycz W., Li Z. 2022. A Two-Stage Approach for Constructing Type-2 Information Granules. IEEE Transactions on Cybernetics. 52(4):2214-2224.

4. Gong Y., Li X., Jiang W. 2020. A New Method for Ranking Discrete Z-number. 2020 Chinese Control And Decision Conference (CCDC). 3591-3596.

5. Hassan S. G., Iqbal S., Garg H., Hassan M., Shuangyin L., Kieuvan T.T. 2020. Designing Intuitionistic Fuzzy Forecasting Model Combined With Information Granules and Weighted Association Reasoning. IEEE Access. Vol. 8. P. 141090-141103.

6. Shan D., Lu W., Yang J. 2019. Interval granular fuzzy models: Concepts and development. IEEE Access. 7:24140–24153.

7. Firsov A.N., Zhilenkov A.A., Chernyi S.G. 2021. Solving problems in transportation systems modeled by the nonlinear Kolmogorov-Feller equation. Journal of Information Technologies and Computing Systems. 2: 84-93.

8. Zhilenkov A.A., Chjornyj S.G. 2014. Primenenie nejronechjotkogo modelirovanija dlja zadach identifikacii mnogokriterial’nosti v transportnoj otrasli [Application of neuro-fuzzy modeling for problems of identification of multicriteria in the transport industry]. Vestnik SamGUPS [JournalSamGUPS]. 1(23):100-106.