Article: A big data-driven framework for enhancing precision level in university ideological and political education: mechanisms and pathways Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.83 pp.60 - 81 Abstract: In university ideological and political education, students' ideological states are dynamic and hard to measure directly. The traditional uniform teaching model cannot achieve individualised instruction, and existing big data research lacks operational precise mechanisms and empirical verification. To address this gap, this paper proposes a hybrid-driven framework integrating fuzzy c-means clustering, random forest, and long short-term memory network. By constructing dynamic student profiles and context-aware recommendation strategies, it enables personalised resource distribution. Experiments show the model increases prediction accuracy from the random forest baseline of 86.2% to 89.5%, and area under the curve from 0.891 to 0.927. The introduced group response consistency index and precise distribution uniformity reach 0.842 and 0.763, respectively, outperforming the uniform distribution strategy without personalised recommendations. This framework fills the empirical gap from theoretical conception to quantifiable deployment. Inderscience Publishers - linking academia, business and industry through research

Title: A big data-driven framework for enhancing precision level in university ideological and political education: mechanisms and pathways

Authors: Mengtian Chen; Zeping Huang

Addresses: School of Marxism, Xinyang Agriculture and Forestry University, Xinyang, 464000, China ' School of Landscape Architecture and Art, Xinyang Agriculture and Forestry University, Xinyang, 464000, China

Abstract: In university ideological and political education, students' ideological states are dynamic and hard to measure directly. The traditional uniform teaching model cannot achieve individualised instruction, and existing big data research lacks operational precise mechanisms and empirical verification. To address this gap, this paper proposes a hybrid-driven framework integrating fuzzy c-means clustering, random forest, and long short-term memory network. By constructing dynamic student profiles and context-aware recommendation strategies, it enables personalised resource distribution. Experiments show the model increases prediction accuracy from the random forest baseline of 86.2% to 89.5%, and area under the curve from 0.891 to 0.927. The introduced group response consistency index and precise distribution uniformity reach 0.842 and 0.763, respectively, outperforming the uniform distribution strategy without personalised recommendations. This framework fills the empirical gap from theoretical conception to quantifiable deployment.

Keywords: big data; ideological and political education; precision education; hybrid learning model.

DOI: 10.1504/IJICT.2026.155122

International Journal of Information and Communication Technology, 2026 Vol.27 No.83, pp.60 - 81

Received: 25 Apr 2026
Accepted: 23 May 2026
Published online: 27 Jul 2026 *

Title: A big data-driven framework for enhancing precision level in university ideological and political education: mechanisms and pathways

Keep up-to-date