A Study on the Precision Effect of Blended Learning in English Teaching and Learning

Abstract

Abstract Realizing personalized and precise teaching using traditional English methods is challenging. This paper proposes a blended precision teaching model that relies on student portraits, using technical methods to provide tailored learning resources for every student. A deep neural network is used to extract student features, and the k-means algorithm is used to construct a clustered portrait of students. Based on the student portraits, the similarity between student objects is calculated, and the collaborative filtering method is combined to achieve personalized recommendations for English learning resources. And the learning warning model is established by considering the ranking order relationship when predicting students’ English scores. Setting up the experimental class and the control class to analyze the effect of blended teaching precision, in terms of English scores, the average score of the experimental class is 6.06 points higher than that of the control class, with a significant P-value of less than 0.05, which shows a considerable difference. Its classroom teaching observation dimension score totaled 91.6, students’ classroom performance and teaching effect performed well, English literacy was improved, and the mean values of each satisfaction of emotional experience showed significant differences (P<0.05). The mean values of several dimensions of learning motivation were higher than those of the control class, with highly significant differences in the dimensions of extrinsic goal orientation, learning beliefs, and intrinsic goal orientation (P<0.01).