Combined Cognitive Model for Forecasting University Activities

Abstract

Purpose of the study. The purpose of the study is to develop a model for predicting university performance indicators based on a cognitive approach, which is based on the construction of a cognitive map that reflects the influence of a set of latent factors on the basic indicators and provides a solution to the problem of scenario forecasting. The degree of achievement of the required values of the basic indicators that determine the ranking of the university depends on the magnitude of the increment of the identified latent factors. The developed model makes it possible to choose the most preferable variant of scenario forecasting under the existing restrictions on the resources allocated for the increment of latent factors.Materials and methods. To achieve this goal, cognitive modeling methods based on gray fuzzy cognitive maps (FCM) were used in combination with methods of interval mathematics and causal algebra. The application of the considered approach made it possible to reduce the uncertainty of expert estimates of the strength of the relationship between the concepts of the cognitive map due to the use of special constructions in the form of interval estimates rather than point estimates when describing the relationships between the concepts, which ensured an increase in the reliability of the modeling results. The developed model is created based on an ensemble of gray FCMs, which, in turn, made it possible to increase the accuracy and reliability of the predictive model. The proposed approach to solving the problem of predicting the activities of the university made it possible to develop an adequate cognitive model.Results. The developed cognitive model of the university’s activities made it possible to analyze the dynamics of changes in factors and their influence on basic indicators, as well as the dynamics of the development of the system of indicators. The calculation made it possible to choose the most cost-effective scenario for incrementing the values of latent factors to obtain the required value of the university ranking in the framework of the QS international institutional ranking of universities. A comparative analysis of the results of scenario forecasting based on conventional FCM, gray FCM, and an ensemble of gray FCM was carried out, which showed the advantage of the proposed approach.Conclusion. During the study, a fuzzy cognitive model was developed for scenario forecasting of measures to achieve the required values of university performance targets in the QS international institutional ranking based on an ensemble of gray FCMs. The developed model provides, under the given constraints, obtaining the most acceptable scenario for planning the increment of basic indicators to target values by identifying the latent factors influencing them and calculating the required values of impulse effects on latent factors.