Research on the Integration Application of Red Culture in Civic and Political Education in Colleges and Universities Based on Multimodal Data Analysis Algorithm

Abstract

Utilizing digital intelligence technology to analyze and process multimodal data to promote the deep integration of red culture and ideological and political education in higher education is an important direction for the current educational reform. This paper constructs a red cultural resource value extraction framework supported by large language model technology, comprising four main modules: red resource integration, knowledge graph construction, knowledge reasoning, and intelligent applications. On the knowledge graph of red cultural resources, a BERT+CRF model is constructed for entity extraction to assist in obtaining graph triples. For personalized learning path resource recommendations, a deep reinforcement learning recommendation algorithm optimized for recommendation diversity is introduced, and its decision-making model is designed. Combining personalized learning path intelligent recommendation theory with the extracted red resources, the framework achieves the recommendation of red cultural resources in college ideological and political education courses. The teaching platform equipped with the recommendation system developed in this paper has received over 80.00% positive evaluations from students regarding its effectiveness and over 85.00% positive evaluations regarding satisfaction in actual applications.