Development Of English Education Using Big Data And Learning Analytics

Abstract

With English education using big data and learning analytics represents a transformative approach to enhancing language learning outcomes and pedagogical practices. By harnessing the power of big data analytics, educational institutions can collect, process, and analyze vast amounts of learner data, including student interactions with digital learning materials, performance on assessments, and engagement metrics. Through sophisticated learning analytics techniques, educators gain valuable insights into students' learning behaviors, preferences, strengths, and areas for improvement. The paper explores the application of big data analytics and learning analytics in English language education within Chinese universities. Utilizing a diverse dataset, including proficiency levels, engagement metrics, and demographic characteristics, the study employs Median Clustering Fuzzy Learning Analytics (MCFLA) to predict students' engagement levels based on language proficiency scores. Findings reveal varying proficiency levels among students, alongside mixed patterns of engagement. The integration of MCFLA proves effective in identifying trends and guiding educational interventions. Median scores provide benchmarks for assessing performance, while demographic factors offer additional insights. Findings reveal varying proficiency levels among students: 40% beginner, 30% intermediate, and 30% advanced. Mixed patterns of engagement are observed, with 50% high engagement, 30% medium, and 20% low. Integration of MCFLA effectively predicts engagement trends, with 75% accuracy. Median scores provide benchmarks for assessing performance: beginner (Vocabulary: 63, Grammar: 66, Reading: 68, Oral Communication: 55), intermediate (Vocabulary: 78, Grammar: 79, Reading: 80, Oral Communication: 72), and advanced (Vocabulary: 90, Grammar: 88, Reading: 95, Oral Communication: 80). Demographic factors such as gender (50% male, 50% female) and socio-economic status (40% middle class, 30% upper class, 30% lower class) offer additional insights.