The rapid growth of multimedia content necessitates powerful technologies to filter, classify, index and retrieve video documents more efficiently. However, the essential bottleneck of image and video analysis is the ...The rapid growth of multimedia content necessitates powerful technologies to filter, classify, index and retrieve video documents more efficiently. However, the essential bottleneck of image and video analysis is the problem of semantic gap that low level features extracted by computers always fail to coincide with high-level concepts interpreted by humans. In this paper, we present a generic scheme for the detection video semantic concepts based on multiple visual features machine learning. Various global and local low-level visual features are systelrtically investigated, and kernelbased learning method equips the concept detection system to explore the potential of these features. Then we combine the different features and sub-systen on both classifier-level and kernel-level fusion that contribute to a more robust system Our proposed system is tested on the TRECVID dataset. The resulted Mean Average Precision （MAP） score is rmch better than the benchmark perforrmnce, which proves that our concepts detection engine develops a generic model and perforrrs well on both object and scene type concepts.展开更多
In this study, methods to classify advertising reviews from shopping mall reviews are suggested. Advertising reviews are mostly written by companies and contain advertising contents. There are a few studies regarding ...In this study, methods to classify advertising reviews from shopping mall reviews are suggested. Advertising reviews are mostly written by companies and contain advertising contents. There are a few studies regarding the classification of opinion spam documents, which is very rare in foreign studies; however, there are no studies that classify advertising reviews from Korean reviews. In this study, the Naive Bayes Classifier was used to classify review documents and the POS （Part-of-Speech）-Tagging and bigram methods were used to extract specific words. The frequency calculation methods for the probability value of specific words were： （1） The general number of appearances of words （2） the frequency calculation of specific words through the suggested Latent Semantic Analysis （LSA）, and by recalculating the result from （1） in （2）, the performances of each method were compared. As a result, the methods from （2） showed 88.43% accuracy which is 8.89% higher than 79.54% which was the previous result from using the POS-Tagging ＋ Bigram method. Therefore, it was proved that the method suggested in this study is effective at classifying or extracting advertising reviews from Korean product review documents.展开更多
基金Acknowledgements This paper was supported by the coUabomtive Research Project SEV under Cant No. 01100474 between Beijing University of Posts and Telecorrrcnications and France Telecom R＆D Beijing the National Natural Science Foundation of China under Cant No. 90920001 the Caduate Innovation Fund of SICE, BUPT, 2011.
文摘The rapid growth of multimedia content necessitates powerful technologies to filter, classify, index and retrieve video documents more efficiently. However, the essential bottleneck of image and video analysis is the problem of semantic gap that low level features extracted by computers always fail to coincide with high-level concepts interpreted by humans. In this paper, we present a generic scheme for the detection video semantic concepts based on multiple visual features machine learning. Various global and local low-level visual features are systelrtically investigated, and kernelbased learning method equips the concept detection system to explore the potential of these features. Then we combine the different features and sub-systen on both classifier-level and kernel-level fusion that contribute to a more robust system Our proposed system is tested on the TRECVID dataset. The resulted Mean Average Precision （MAP） score is rmch better than the benchmark perforrmnce, which proves that our concepts detection engine develops a generic model and perforrrs well on both object and scene type concepts.
文摘In this study, methods to classify advertising reviews from shopping mall reviews are suggested. Advertising reviews are mostly written by companies and contain advertising contents. There are a few studies regarding the classification of opinion spam documents, which is very rare in foreign studies; however, there are no studies that classify advertising reviews from Korean reviews. In this study, the Naive Bayes Classifier was used to classify review documents and the POS （Part-of-Speech）-Tagging and bigram methods were used to extract specific words. The frequency calculation methods for the probability value of specific words were： （1） The general number of appearances of words （2） the frequency calculation of specific words through the suggested Latent Semantic Analysis （LSA）, and by recalculating the result from （1） in （2）, the performances of each method were compared. As a result, the methods from （2） showed 88.43% accuracy which is 8.89% higher than 79.54% which was the previous result from using the POS-Tagging ＋ Bigram method. Therefore, it was proved that the method suggested in this study is effective at classifying or extracting advertising reviews from Korean product review documents.
