Research And Application Of Teaching Video Retrieval Based On Clustering

The development of video production and propagation velocity has been promoted greatly owning to the progress of Internet technology, and streaming media video data on the Internet is growing at an alarming rate. It has become a trend that video in the form of streaming media shared by network users. Video, as a type of resource in education and teaching, has become an important application form of distance teaching with the rapid development of network technology and the increasingly common used streaming video. However, with the increase of video data, people fell into an embarrassing situation that video data is very rich but it is hard to find a satisfactory one. Scholars and research institutions both home and abroad have done a great deal of work on how to quickly get the video that meet the needs, so that they promoted the research of content-based video retrieval method and related technology.On the basis of the research of current video retrieval technology and text mining technique, the author put forward the framework thought based on the text clustering teaching video system. According to this thought, the author mainly carries on the following two aspects of work.(1) To design and implement a Chinese text clustering based experiment based on classical K mean. The author describes in details the text pretreatment process, the experimental process and the experimental verification process, and introduces with emphasis the segmentation of Chinese word, the dimension reduction processing work of text feature, the algorithm thought of classical K-means and the process of experimental data analysis. By means of the experimental data to prove the experiments, the feasibility and reliability of the experiments are good.(2) On the basis of the feasibility verification of the above experiments, then having analyzed the characteristics of teaching video and obtained the video content and analyzed the video retrieval system and its related technologies work, the retrieval model of teaching video has been established based on the text clustering.In this model, the author uses video text as the index of both retrieval and the optimization of its result, and then makes an optimizing classification study on the query result. Finally, taking a course teaching video as a case to make tests, the author gets a more satisfactory result.