Functional Classification of Artificial Intelligence Applications within Learning Management Systems: An Analytical Study of Concepts and Applications

Abstract

This paper presents a conceptual and classificatory analytical study aimed at constructing a functional taxonomy of artificial intelligence applications within Learning Management Systems (LMS) during the period 2020–2025. The study responds to the problem of conceptual confusion and terminological overlap between artificial intelligence, learning analytics, rule-based automation, and personalization.

The paper adopts a structured review of evidence and codes “functional features” as the unit of observation into defined categories according to operational definitions and attribution rules based on the “traceable end output” within the LMS environment.

This approach produces a single classification table comprising five functional categories: prediction, recommendation, automated assessment, conversational support, and administrative decision support. Each category is linked to expected data inputs and observable outputs, along with explicit rules for managing overlap when more than one function appears within a single feature.

The paper concludes that regulating attribution through traceable outputs reduces terminological ambiguity and improves the comparability and descriptive evaluation of artificial intelligence functions within LMS environments, without making causal claims regarding educational impact.