Feiliang Ren

Chinese-English Organization Name Translation Based on Correlative Expansion

This paper presents an approach to translating Chinese organization names into English based on correlative expansion. Firstly, some candidate translations are generated by using statistical translation method. And several correlative named entities for the input are retrieved from a correlative named entity list. Secondly, three kinds of expansion methods are used to generate some expanded queries. Finally, these queries are submitted to a search engine, and the refined translation results are mined and re-ranked by using the returned web pages. Experimental results show that this approach outperforms the compared system in overall translation accuracy.

Translate Chinese organization names using examples and web

This paper proposes a new approach for translating Chinese organization names that uses example-based method along with web assistance. It consists of two phases, first, it generates a translation candidate for the input Chinese organization name by an example-based translation method; and secondly, it uses the web to amend this translation candidate so as to finish such tasks: translation candidate reordering, word selection revising, and adjustment of the use of function words. Experimental results show that our method outperforms competing traditional statistical translation method in the task of translating Chinese ONs.

Web-based technical term translation pairs mining for patent document translation

This paper proposes a simple but powerful approach for obtaining technical term translation pairs in patent domain from Web automatically. First, several technical terms are used as seed queries and submitted to search engineering. Secondly, an extraction algorithm is proposed to extract some key word translation pairs from the returned web pages. Finally, a multi-feature based evaluation method is proposed to pick up those translation pairs that are true technical term translation pairs in patent domain. With this method, we obtain about 8,890,000 key word translation pairs which can be used to translate the technical terms in patent documents. And experimental results show that the precision of these translation pairs are more than 99%, and the coverage of these translation pairs for the technical terms in patent documents are more than 84%.

A dynamic weighted method with support vector machines for Chinese word segmentation

We explore how a dynamic weighted method with SVM works well for a Chinese word segmentation task. We set up two systems, System1 uses the uniform weight and we take it as our baseline system, System1 uses the dynamic weighted method as we proposed. We compare the performance of the two systems under different experiments conditions, and experiments results show that System2 got an outstanding performance in every condition we tested. It indicates that the dynamic weighted method we proposed has a stronger performance and can be used in many other SVM task such as chunking, POS and so on. At last, we describe a trick that can calculate the segmentation precision ratio and recall ratio for a segmentation task as soon as the end of the training or test process of a SVM procedure at essentially no extra cost.

BiTCNN: A Bi-Channel Tree Convolution Based Neural Network Model for Relation Classification.

Relation classification is an important task in natural language processing (NLP) fields. State-of-the-art methods are mainly based on deep neural networks. This paper proposes a bi-channel tree convolution based neural network model, BiTCNN, which combines syntactic tree features and other lexical level features together in a deeper manner for relation classification. First, each input sentence is parsed into a syntactic tree. Then, this tree is decomposed into two sub-tree sequences with top-down decomposition strategy and bottom-up decomposition strategy. Each sub-tree represents a suitable semantic fragment in the input sentence and is converted into a real-valued vector. Then these vectors are fed into a bi-channel convolutional neural network model and the convolution operations re performed on them. Finally, the outputs of the bi-channel convolution operations are combined together and fed into a series of linear transformation operations to get the final relation classification result. Our method integrates syntactic tree features and convolutional neural network architecture together and elaborates their advantages fully. The proposed method is evaluated on the SemEval 2010 data set. Extensive experiments show that our method achieves better relation classification results compared with other state-of-the-art methods.

Embedding Syntactic Tree Structures into CNN Architecture for Relation Classification

Relation classification is an important task in natural language processing (NLP) fields. State-of-the-art methods are mainly based on deep neural networks. This paper proposes a new convolutional neural network (CNN) architecture which combines the syntactic tree structure and other lexical level features together for relation classification. In our method, each word in the input sentence is first represented as a k-size word sequence which contains the context information of the considering word. Then each of such word sequence is parsed into a syntactic tree structure and this kind of tree structure is further mapped into a real-valued vector. Finally, concatenated with the attention features for the words among the marked entities, all of these features are fed into a CNN model for relation decision. We evaluate our method on the SemEval 2010 relation classification task and experimental results show that our method outperforms previous state-of-the-art methods under the condition of without using external linguistic resources like WordNet.

A novel Chinese-English on translation method using mix-language web pages

In this paper, we propose a novel Chinese-English organization name translation method with the assistance of mix-language web resources. Firstly, all the implicit out-of-vocabulary terms in the input Chinese organization name are recognized by a CRFs model. Then the input Chinese organization name is translated without considering these recognized out-of-vocabulary terms. Secondly, we construct some efficient queries to find the mix-language web pages that contain both the original input organization name and its correct translation. At last, a similarity matching and limited expansion based translation identification approach is proposed to identify the correct translation from the returned web pages. Experimental results show that our method is effective for Chinese organization name translation and can improve performance of Chinese organization name translation significantly.

An Effective Approach for Coreference Resolution

We present a machine learning approach for coreference resolution of noun phrases. In our method, we use CRFs as a basic training model, and use active learning method to generate combined features so as to use existing features more effectively. We also propose a novel clustering algorithm which uses both linguistic knowledge and statistical knowledge. We build a coreference resolution system based on the proposed method and evaluate its performance from three aspects: the contributions of active learning; the effects of different clustering algorithms; and the resolution performance of different kinds of NPs. Experimental results show that additional performance gain can be obtained by using active learning method; clustering algorithm has a great effect on coreference resolutions performance and our clustering algorithm is very effective; and the key of coreference resolution is to improve the performance of the normal nouns resolution, especially the pronouns resolution.