A Comprehensive Model for Recommending Personalized Learning Resources for the Development of Linguistic Competence

Abstract

With the continuous advancement of globalization and informatization, the linguistic competence of college students has become a key index for evaluating their comprehensive quality. Faced with diverse needs of students and educational environments, it is increasingly important and complex to accurately locate the linguistic competence goals of college students. Although existing research methods, such as standardized testing and teacher assessment, provide certain convenience, they rely on single data sources and have a certain degree of subjectivity, which limits their universality and accuracy. This study aimed to solve this problem by doing comprehensive research on two aspects: first, curriculum analysis based on relation extraction. A relation extraction model, such as Casrel, was used for advanced text analysis, which provided educators with more in-depth insights; second, personalized learning material recommendation based on text recommendation. Personalized learning paths were provided for students of different levels using the abstractness-based text recommendation algorithm. This study not only filled the gaps in existing research methods, but also provided a new, scientific and efficient solution, helping improve the quality of education and promote the formulation of scientific education policies.