Fusion of Machine Learning for Teaching Case Research on Algorithm Course

Abstract

“Algorithm Design and Analysis” is a professional compulsory course for computer undergraduates. A solid grasp of the content of the course is of great importance for students to engage in relevant positions after graduation such as algorithm engineers or to further study. Nowadays, a large number of application problems need to be solved by machine learning algorithms. In fact, the underlying implementation details of many machine learning algorithms are also derived from these basic algorithms. However, there are few descriptions related to machine learning algorithms in current algorithm courses. In the process of teaching basic algorithms, if automatic association with machine learning algorithms can be realized, knowledge will root in the core knowledge base of students, thereby realizing continuous extension of knowledge. Based on this, this paper effectively associates divide and conquer and greedy algorithms with the machine learning representative --- decision tree algorithm, so as to improve the students' analogy of relevant contents and knowledge and draw inferences from one instance.