About the System of Profiling User Actions Based on the Behavior Model

Autor: Marina N. Zhukova, Yegor V. Anashkin
Rok vydání: 2021
Předmět:
Zdroj: 2021 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (ElConRus).
Popis: The paper considers the issue of increasing the level of trust to the user of the information system by applying profiling actions. The authors have developed the model of user behavior, which allows to identify the user by his actions in the operating system. The model uses a user's characteristic metric instead of binary identification. The user's characteristic demonstrates the degree to which the current actions of the user corresponding to the user's behavior model. To calculate the user's characteristic, several formulas have been proposed. The authors propose to implement the developed behavior model into the access control model. For this purpose, the authors create the prototype of the user action profiling system for Windows family operating systems. This system should control access to protected resources by analyzing user behavior. The authors performed a series of tests with this system. This allowed to evaluate the accuracy of the system based on the proposed behavior model. Test results showed the type I errors. Therefore, the authors invented and described a polymodel approach to profiling actions. Potentially, the polymodel approach should solve the problem of the accuracy of the user action profiling system.
Databáze: OpenAIRE