Core architecture for cooperation between smart tutors

Core architecture for cooperation between smart tutors
(Project Manager: Dr. Omar Álvarez Xochihua, Ph.D.)

This research addresses the field of educational technology, through the design, implementation and evaluation of a software architecture that allows communication between intelligent tutors (pedagogical agents), allowing to homogenize the student model to be shared between intelligent tutors from different domains. Intelligent tutoring systems (ITS, for its acronym in English), are virtual advisors based on software, which arise as an option to increase the academic performance of students by providing them personalized advice; through knowledge previously obtained from human advisers.

For knowledge modeling, semantic web tools were used; the decision was made to use knowledge modeling techniques through ontological semantics. The design of the integral architecture of intelligent tutors was based on a metamodel based on Model Driven Architecture, separating the functional flows (services) from the technologies that implement them. At the same time, a deliberative multi-agent communication architecture is sustained, through a positive relationship and implicit interaction. The communication protocol between tutors implemented is carried out using the xAPI specification and an LRS repository, due to the benefits it offers so that very different systems can communicate with each other while interacting with different technologies and content.