The Development of the Staking-Ensemble of Methods for Analyzing Academic Data

Abstract

Within the framework of this article, the Staking-ensemble of methods for analyzing academic statistics is proposed, which makes it possible to increase the effectiveness of implementing academic monitoring tasks. The relevance of the topic is the need to develop a distributed information and analytical system that integrates information resources and general principles of models and methods of monitoring the infrastructures of academic facilities based on education statistics. Education statistics is a system of indicators characterizing quantitative and qualitative changes taking place in the field of education which makes it possible to obtain information for each level of education on the number of educational institutions, the contingent of students, characteristics of the internal efficiency of the learning process, data on admission to educational institutions, graduation of specialists, quantitative and qualitative characteristics of the teaching staff, the state of the material and technical base of educational institutions. The object of the study is the system of formation of statistical data in education. The subject of the study are the approaches of combining intelligent methods of data analysis. The idea of the work is the use of modern methods of data processing in the implementation of academic monitoring in order to successfully solve the tasks of the state program for the development of education. The goal of the study is to develop an algorithm for Staking-ensemble methods for analyzing academic statistics to improve the effectiveness of academic monitoring tasks. The scientific novelty of the research is the staking-ensemble for analyzing education statistics, which aggregated the following 3 types of intellectual models: the Bayes algorithm, the decision tree algorithm, and the neural network. The practical importance of the research results lies in the applicability of the proposed Staking-ensemble algorithm for solving the problems of information and analytical support of management decision-making when tracking the business processes of monitoring, controlling and forecasting the state of distributed academic objects at various levels of training, management and functioning.