We introduce a novel Sermntic-Category- Tree （SCT） model to present the sen-antic structure of a sentence for Chinese-English Machine Translation （MT）. We use the SCT model to handle the reordering in a hierarchic...We introduce a novel Sermntic-Category- Tree （SCT） model to present the sen-antic structure of a sentence for Chinese-English Machine Translation （MT）. We use the SCT model to handle the reordering in a hierarchical structure in which one reordering is dependent on the others. Different from other reordering approaches, we handle the reordering at three levels： sentence level, chunk level, and word level. The chunk-level reordering is dependent on the sentence-level reordering, and the word-level reordering is dependent on the chunk-level reordering. In this paper, we formally describe the SCT model and discuss the translation strategy based on the SCT model. Further, we present an algorithm for analyzing the source language in SCT and transforming the source SCT into the target SCT. We apply the SCT model to a role-based patent text MT to evaluate the ability of the SCT model. The experimental results show that SCT is efficient in handling the hierarehical reordering operation in MT.展开更多
There is a major defect when using the traditional topic-opinion model for post opinion classifications in an online forum discussion.The accuracy of the classification based on the topic-opinion model highly depends ...There is a major defect when using the traditional topic-opinion model for post opinion classifications in an online forum discussion.The accuracy of the classification based on the topic-opinion model highly depends on the observable topic-opinion features aiming at the subject,while a large number of posts do not have such features in a forum.Therefore,for the most part,the accuracy is less than 78%.To solve this problem,we propose a new method to identify post opinions based on the Tree Conditional Random Fields(T-CRFs)model.First,we select the topic-opinion features of the posts and associated opinion features between posts to construct the T-CRFs model,and then we use the T-CRFs model to label the opinions of the tree-structured posts under the same topic iteratively to reach a maximum joint probability.To reduce the training cost,we design a simplified tree diagram module and some feature templates.Experimental results suggest the proposed method costs less training time and improves the accuracy by 11%.展开更多
基金supported by the National High Technology Research and Development Program of China under Grant No.2012AA011104the Fundamental Research Funds for the Center Universities
文摘We introduce a novel Sermntic-Category- Tree （SCT） model to present the sen-antic structure of a sentence for Chinese-English Machine Translation （MT）. We use the SCT model to handle the reordering in a hierarchical structure in which one reordering is dependent on the others. Different from other reordering approaches, we handle the reordering at three levels： sentence level, chunk level, and word level. The chunk-level reordering is dependent on the sentence-level reordering, and the word-level reordering is dependent on the chunk-level reordering. In this paper, we formally describe the SCT model and discuss the translation strategy based on the SCT model. Further, we present an algorithm for analyzing the source language in SCT and transforming the source SCT into the target SCT. We apply the SCT model to a role-based patent text MT to evaluate the ability of the SCT model. The experimental results show that SCT is efficient in handling the hierarehical reordering operation in MT.
基金supported by the National Natural Science Foundation of China under Grant No. 60873246China Information Technology Security Evaluation Centre
文摘There is a major defect when using the traditional topic-opinion model for post opinion classifications in an online forum discussion.The accuracy of the classification based on the topic-opinion model highly depends on the observable topic-opinion features aiming at the subject,while a large number of posts do not have such features in a forum.Therefore,for the most part,the accuracy is less than 78%.To solve this problem,we propose a new method to identify post opinions based on the Tree Conditional Random Fields(T-CRFs)model.First,we select the topic-opinion features of the posts and associated opinion features between posts to construct the T-CRFs model,and then we use the T-CRFs model to label the opinions of the tree-structured posts under the same topic iteratively to reach a maximum joint probability.To reduce the training cost,we design a simplified tree diagram module and some feature templates.Experimental results suggest the proposed method costs less training time and improves the accuracy by 11%.
