Self-adaptation of e-learning software based on observing the information environment

Abstract

Purpose of the research. The purpose of the study is to develop a new universal method of self-adaptation of applied software systems used in the field of e-learning (E-Learning). Self-adaptation refers to the ability of a software application to change its own structure and behavior depending on external circumstances, which include, for example, the trainee’s personal characteristics, which is especially important for systems used in education. Such self-adaptive behavior should be sufficiently flexible and not be reduced to the choice of one of the many behavioral options predetermined by the developer (such behaviors should also be generated throughout the system’s life cycle). Materials and methods. The method being developed uses an array of user reviews about software as initial data, for the subsequent processing of which the methods of latent-semantic and distributive-statistical analysis are used. To represent the generalized self-adaptive structure of the system, models of characteristics are used. The configuration of the model of characteristics is a separate state of the self-adaptive system, they are generated automatically during the program’s life cycle as follows: based on an array of user reviews, a semantic network of basic concepts characterizing the program is formed, which is further compared with the original model of characteristics and personal characteristics of the user who left review. Determining a user’s personal characteristics can be done in a variety of ways (for example, using psychological testing or by analyzing learning outcomes). Results. The main results of the study are: 1) universal principles of building a self-adaptive e-learning system 2) a way of presenting the self-adaptive structure of a software system in the form of a characteristics model relevant to a wide range of software 3) a new universal method of self-adapting applied software used in E-Learning the main differences of which from the existing ones are, firstly, in using the opinions of the users of the system themselves to adjust with self-adaptive behavior, secondly, in the possibility of generating new states of the system throughout the entire period of its operation. Conclusion. The developed theoretical apparatus makes it possible to significantly individualize the learning process, take into account the opinions and inclinations of the students themselves, reduce the role of the pedagogical worker in the assessment of knowledge and skills. In addition to problems of a purely educational nature, the application of the method also allows you to successfully resolve technical issues related to the development of software in general. Such problems include, for example, the problem of software complexity, when a program that shows good results in some operating conditions shows insufficient performance in others. Also a serious task, which the proposed method can cope with, is the task of increasing the life cycle of a software system.