Innovative Teaching Via Sustainable Vocational Education with an Improved Ant Colony Algorithm

Abstract

Although students’ test scores provide an important reference for teaching and learning, research scholars still need to objectively analyze the scores. Under the current situation where English performance of vocational education students does not achieve satisfactory results, this research uses a clustering algorithm to improve on the ant colony optimization algorithm. This ant colony clustering analysis algorithm is improved by incorporating two optimization strategies, and the test scores of vocational education students are introduced as the original data for cluster analysis. The optimal number of ant colonies is nine, when the three error values of the two ant colony algorithms are minimized. The convergence values of the three ant colony algorithms are smallest when there are 200 training cycles or when the training batch size is 1000, resulting in upgraded ant colony clustering algorithm convergence values of 0.498 and 1.523, respectively. The performance of the student evaluation model combined with the ant colony clustering optimization algorithm improved, followed by CF, FOA, and BP. KNN had the worst performance. Data mining on student performance can be done via research that can provide specialized advice on students