<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>The trend of our country into an aging society is becoming more and more obvious, and the elderly education service has become the focus of social attention. The integration of artificial intelligence into elderly education is one of the inspiring ideas in which the personalized recommendation algorithm can recommend educational resources according to the characteristics of older people, and it has the prospect of application. In this paper, we first provide a CTransD-GAT recommendation model based on a knowledge graph, which improves traditional problems such as data sparsity by setting weight preference and feature aggregation. A dynamic preference-capturing method is proposed based on contextual interaction to capture the variable user learning interests more accurately and flexibly. This paper examines the practical utility of personalized recommendation methods for educational resources based on these two improved techniques. The post-test mean score of each knowledge module test of the experimental group is improved by 1.83 points compared with the pre-test, 11.46 points improve the score of teaching ability, and the scores of perceived usefulness, ease of use, and intention to use are 3.82, 3.89, and 3.97, respectively. It shows that the improved educational resource recommendation model has an excellent effect on improving knowledge structure and teaching ability, and it is characterized by simplicity and ease of use.</p>
</abstract>

The trend of our country into an aging society is becoming more and more obvious, and the elderly education service has become the focus of social attention. The integration of artificial intelligence into elderly education is one of the inspiring ideas in which the personalized recommendation algorithm can recommend educational resources according to the characteristics of older people, and it has the prospect of application. In this paper, we first provide a CTransD-GAT recommendation model based on a knowledge graph, which improves traditional problems such as data sparsity by setting weight preference and feature aggregation. A dynamic preference-capturing method is proposed based on contextual interaction to capture the variable user learning interests more accurately and flexibly. This paper examines the practical utility of personalized recommendation methods for educational resources based on these two improved techniques. The post-test mean score of each knowledge module test of the experimental group is improved by 1.83 points compared with the pre-test, 11.46 points improve the score of teaching ability, and the scores of perceived usefulness, ease of use, and intention to use are 3.82, 3.89, and 3.97, respectively. It shows that the improved educational resource recommendation model has an excellent effect on improving knowledge structure and teaching ability, and it is characterized by simplicity and ease of use.

Research on Practical Innovation of Elderly Education Service System in the Era of Artificial Intelligence

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

The trend of our country into an aging society is becoming more and more obvious, and the elderly education service has become the focus of social attention....

Research on Practical Innovation of Elderly Education Service System in the Era of Artificial Intelligence

Data publikacji: 05 sie 2024

Zakres stron: -

Otrzymano: 23 mar 2024

Przyjęty: 16 cze 2024

DOI: https://doi.org/10.2478/amns-2024-2071

Słowa kluczowe
Personalized recommendation algorithm, CTransD-GAT, Dynamic preference capture, Geriatric education

© 2024 Xiao Li, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Research on Practical Innovation of Elderly Education Service System in the Era of Artificial Intelligence

Data publikacji: 05 sie 2024

Zakres stron: -

Otrzymano: 23 mar 2024

Przyjęty: 16 cze 2024

DOI: https://doi.org/10.2478/amns-2024-2071

Słowa kluczowePersonalized recommendation algorithm, CTransD-GAT, Dynamic preference capture, Geriatric education

© 2024 Xiao Li, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Słowa kluczowe
Personalized recommendation algorithm, CTransD-GAT, Dynamic preference capture, Geriatric education