The value and practical path of promoting the innovative thought of higher education under the background of deep learning

Abstract

Abstract The value and practical path of how to promote innovative ideas in higher education in the context of deep learning is a hot issue of concern to many scholars. The article explores the execution steps of this recommendation system through the basic theoretical study of the collaborative filtering recommendation algorithm and further proposes a MapReduce-based collaborative filtering algorithm that utilizes Hadoop to run the ItemCF algorithm's steps in order to complete the Map/Reduce job. The article concludes by systematically testing the algorithm and applying it to evaluate specific innovative ideas in higher education. After the system comparison test, it is found that the recommendation algorithm proposed in this article is better than the KNN collaborative filtering algorithm in the same proportion of the test set, and the memory occupancy, CPU occupancy, and the average response time of the interface are all within the expected value. In the paired-sample t-test analysis of the dimensions of the pre-and post-test data of the values of the students in the experimental class, the P-values of the six dimensions of patriotic sentiment, scientific spirit, humanistic spirit, aesthetic consciousness, ideology and morality, and cooperative spirit are all less than 0.05, which is a significant difference.