作者: José M. Perea-Ortega , Arturo Montejo-Ráez , M. Teresa Martín-Valdivia , L. Alfonso Ureña-López
DOI: 10.1007/S10844-010-0123-6
关键词:
摘要: In this paper, several experiments about video categorization using a supervised learning approach are presented. To end, the VideoCLEF 2008 evaluation forum has been chosen as experimental framework. After an analysis of corpus, it was found that transcriptions not best source information in order to identify thematic streams. Therefore, two web-based corpora have generated aim adding more informational sources by integrating documents from Wikipedia articles and Google searches. A number test data accomplished. Several machine algorithms proved validate effect corpus on final results: Naive Bayes, K-nearest-neighbors (KNN), Support Vectors Machine (SVM) j48 decision tree. The results obtained show web can be useful for generating classification models data.