I. Introduction
Due to the rapid growth of sci & tech and information technology (IT), mankind has entered the information age. In the 21st century, with the highly developed and continuous perfection of Internet technology and communication technology, the speed of human beings generating and updating knowledge is getting faster and faster [1]. Therefore, higher requirements are put forward for the way of knowledge acquisition. The video instructional system based on traditional C/S mode brings convenience to distance education, but there are also some shortcomings, such as high construction cost, heavy server load, high demand for network bandwidth and poor playback quality when there are too many users [2–3]. In P2P streaming media system, a subset of nodes own a certain media file, and they stream the media file to the requesting node [4]. On the other hand, the requesting nodes will play back and store the media data during the media streaming session, and they will also become the supplying nodes of the media files after the streaming session [5]. An important issue in college teaching is the quality of teaching, and how decision makers can better evaluate and improve teaching management is a very important issue [6]. At present, various types of information management systems have been established in various departments of universities, and their development and application have greatly improved work efficiency [7]. Clustering is to divide data manipulation objects into multiple classes or clusters, and combine them according to their similar characteristics to form many categories, and the quantity of categories is determined according to the actual situation [8–9]. As an important technology in data mining, cluster analysis is more and more widely used in database field. This article integrates online video instructional resources based on clustering algorithm, and builds online video instructional system in universities.