Article: Application of intelligent personalised information recommendation technology in the operation of new media platform Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.34 pp.24 - 44 Abstract: The diversity and complexity of new media information bring great pressure to personalised information recommendation, so intelligent terminals need to be improved with the support of reliable recommendation algorithms. This paper proposes a time-aware (TA) multi-hop path recommendation inference model TACKG-TDPRec to improve the intelligent push effect of personalised information on new media platforms, and introduces the TA path diversity reasoning method, uses time information to improve the accuracy of recommendation results, and enhances personalised diversity rewards on the basis of personalised diversity rewards designed according to user needs. It can be seen that the AUC of the recommended model is as high as 97.94, which is much higher than other models of the same type. From the diversity comparison, it can be seen that TACKG-TDPRec model can adapt to various types of information recommendation needs, and the similarity of items is low, so it has strong practicability. Inderscience Publishers - linking academia, business and industry through research

Title: Application of intelligent personalised information recommendation technology in the operation of new media platform

Authors: Cheng Shaoxiao

Addresses: School of Communication, Nanchang Institute of Technology, Nanchang, China

Abstract: The diversity and complexity of new media information bring great pressure to personalised information recommendation, so intelligent terminals need to be improved with the support of reliable recommendation algorithms. This paper proposes a time-aware (TA) multi-hop path recommendation inference model TACKG-TDPRec to improve the intelligent push effect of personalised information on new media platforms, and introduces the TA path diversity reasoning method, uses time information to improve the accuracy of recommendation results, and enhances personalised diversity rewards on the basis of personalised diversity rewards designed according to user needs. It can be seen that the AUC of the recommended model is as high as 97.94, which is much higher than other models of the same type. From the diversity comparison, it can be seen that TACKG-TDPRec model can adapt to various types of information recommendation needs, and the similarity of items is low, so it has strong practicability.

Keywords: new media; personalisation; information; intelligent push.

DOI: 10.1504/IJICT.2025.148819

International Journal of Information and Communication Technology, 2025 Vol.26 No.34, pp.24 - 44

Received: 12 May 2025
Accepted: 01 Jul 2025
Published online: 26 Sep 2025 *

Title: Application of intelligent personalised information recommendation technology in the operation of new media platform

Keep up-to-date