With the in-depth application of Internet technology in the field of English education, the connection between Internet technology and the reverse dialing effect of English teaching in colleges and universities has become increasingly close. The theory of the reverse dialing effect is introduced in this paper, and an Internet technology framework is proposed that incorporates an online English learning platform and a personalized recommendation model of English resources. The viscous binary particle swarm algorithm with nonlinear factors is used as the algorithm for recommending learning paths to build the online learning platform. At the same time, the online learning platform is also used as the background, and the S-CNN model is constructed by relying on the convolutional neural network model framework to propose a personalized course resource recommendation algorithm based on deep learning. University Q in Changchun City, Jilin Province, China, was used as the research site for practice. The English scores of the students in the experimental class were significantly improved, and the mean values of cooperative communication ability, problem-solving ability, and information literacy in English learning ability were higher than those of the control class by 4.427, 4.394 and 3.918, respectively, which showed significant differences (P<0.05). The average value for all learning status dimensions after the experiment was greater than 3.5, which was higher than before the experiment. The mean value of English teachers’ total teaching beliefs reached 163.76, and the mean values of all classroom behaviors except feedback and guidance behaviors were all higher than 3.5. The positive effects of students’ learning and teachers’ teaching were significantly increased.