Application and Effect Analysis of Intelligent Algorithms in College English Blended Teaching System

Blended teaching is hybrid teaching, combining traditional face-to-face instruction with digital learning methods to create a more flexible and effective educational experience. This system offers a personalized and dynamic learning environment to access materials during face-to-face interactions with peers and professors. The research aims to develop the effect analysis of intelligent algorithms in the English blended learning system for college. The research proposes a novel Mayfly Optimized Efficient Recurrent Neural Network (MO-ERNN) model for predicting student performance and enhancing teaching effectiveness. The research was conducted with 168 college students enrolled in a teaching system, divided into two groups: the experimental group received the intelligent blended teaching system, and the control group received the standard teaching approach without access to the intelligent system. Standardized pre- and post-tests covering vocabulary, grammar, reading comprehension, and listening skills were utilized to assess English competence and academic progress in both groups. The IBM SPSS software 29 was used to implement. The pre- and post-test results were compared using statistical techniques such as regression analysis, paired [Formula: see text]-tests, and analysis of variance (ANOVA). With a statistically significant [Formula: see text]-value of 0.0001, the experimental group outperformed the control group in terms of scores. Lesson facilitation performance scores derived from classroom observations indicated that the experimental group had improved English literacy skills, students’ participation, and general teaching effectiveness. The proposed method MO-ERNN attained accuracy (93%), precision (90%), recall (88%), and F1-score (85%). The research focuses on the intelligence of the College English Blended Learning System, and it has displayed the potential to enhance the performance of students.