Personalized Learning Path of Foreign Language for College Students by K-means clustering and T-test methods

This paper primarily explores the personalized learning path of foreign language learners based on educational big data and learner profiles. We collected data on English learning behaviors and quantitative survey results from 299 non-English major university students at a university in Hangzhou, which were later analyzed and categorized across four dimensions: foreign language learning motivation, foreign language learning style, foreign language learning status, and comprehensive foreign language proficiency. We use K-means clustering and T-test methods to analyze data by SPSS. Significant distinctions were observed among learners across these dimensions. Consequently, for labeled and personalized learner groups, it is recommended to design and intervene in teaching from the perspectives of identifying learners’ interests, catering to diverse learning styles, fully utilizing early warning systems, and adjusting learning pace and content. These strategies aim to help foreign language learners improve their foreign language proficiency at the 1% significance level.