Web Video Classification Based On Social Information

With the rapid development of web technology, Internet has turned into a famous information shared platform for users. As a carrier of text, image and sound, web videos are popular online communication media. Because the number of web videos increases steadily in the Internet, web videos classification becomes a cutting edge research area and proves to be a challenge task.Web video categorization is a fundamental task for web video search. In this thesis, we explore web video categorization from a new perspective, by integrating the model-based and data-driven approaches to boost the performance. The boosting comes from two aspects: one is the performance improvement for text classifiers through query expansion from related videos and user videos. The model-based classifiers are built based on the text features extracted from title and tags. Related videos and user videos act as external resources for compensating the shortcoming of the limited and noisy text features. Query expansion is adopted to reinforce the classification performance of text features through related videos and user videos. The other improvement is derived from the integration of model-based classification and data-driven majority voting from related videos and user videos. We explore the user interest and related videos information, using majority voting method combine semantic result from SVM classification to improve classification result. Semantic meaning from text, video relevance from related videos, and user interest induced from user videos, are combined to robustly determine the video category. Their combination from semantics, relevance and interest further improves the performance of web video categorization. Experiments on YouTube videos demonstrate the significant improvement of the proposed approach compared to the traditional text based classifiers.In the process of web video classification, we confront with managing huge numbers of classification data. We use Hadoop distributed platform to handle mass data pre-processing. Hadoop has been successfully used in famous Internet companies such as Yahoo, Facebook, and is a key technology in cloud computing. We analyze the structure of Hadoop platform and programming framework and implement huge numbers of data preprocessing in Hadoop platform. We research the parallel arithmetic process and implement the interface for image processing; We compare the numerous data processing result from PC and Hadoop platform. Finally, we analyze time efficiency.