2000 Volume 14 Issue 5
Published: 16 October 2000
-
Select all|
-
Review2000, 14(5): 1-7,15.This paper proposes a word sense disambiguating model ,which is based on the word’s semantic context . The model utilizes the knowledge dictionary that reflects the semantic concurrence information of English words. And the correlation of word semantic marks is obtained through statistical analysis of large real corpus.By evaluating each possible semantic context word ,the most reasonable sense of the sense ambiguous word is selected. The experiment of this method proves an increase of 10% in precision.
-
Review2000, 14(5): 8-15.There is now a common consensus in the field of Machine Translation (MT) that a well-qualified system should integrate both language-specific linguistic knowledge and language-independent world knowledge. Recently , ontologies are widely used to model world knowledge in knowledge engineering. This paper introduces an ontology-based English-Chinese MT system ,where ,an ontology was developed to model world knowledge. It has been built partly by organizing concepts into a hierarchy and connecting their internal structures into a network.By mapping words in a specific language to concepts in the ontology ,the system can support disambiguation in source language analysis and words chosen in target language generation. Concepts in the ontology can be also acted as concepts in interlinguas which are represented by Conceptual Graph in our system.
-
Review2000, 14(5): 16-22.The verbal valence can clearly show the essential complements in a sentence. So we can translate these parts according to the verbal valence with high quality.But the descriptive components are so different as the essential parts ,which are more free in a sentence ,that we cann’t apply the method based on the verbal valence to translate them. In this paper ,the translating of descriptive parts are discussed in detail.
-
Review2000, 14(5): 23-30.I-tree was published in 1989. It was based upon the ‘three head hypothesis’ proposed by the Chinese famous linguist Li-Jing-xi. It possesses the universal grammar characters. Since then I-Tree has been rapidly developed and matured. LFG was published in 1982. It is the most challenging grammar among the grammars. The challenging part of LFG is the f-structure ,which is very similar to I-Tree. However it is less simple ,straight ,easily read and put into practice.
-
Review2000, 14(5): 31-36.In this paper ,a text-independent speaker verification method based on the information of candidate sequences of speech segment-identification is described. And the influences caused by the changes of the key factors in the method such as the number of the cluster centers ,the threshold and the judge rules are analyzed emphatically. Through experiments it has been proved that the information of candidate sequences of speech segment-identification works very well in speaker verification and it is an effective auxiliary method in improving the performance of speaker-verification system. Finally ,the deficiency of the new method is pointed out and future work is depicted.
-
ReviewA Multiple Classifiers Integrated System of Online Recognizing Natural Handwritten Chinese Character2000, 14(5): 37-41.In the paper ,a new multiple classifiers integrated model of online recognizing natural handwritten Chinese character is presented. By integrating three independent classification algorithms based on 01-code ,WB-code and SO-code ,combining various features into classifiers and recognizing HanZi on pattern different structure layer ,our system comes at a correct recognition rate of 98.6 per cent .
-
Review2000, 14(5): 42-48.The feature vectors of handwritten Chinese characters are often more than several hundred. In such a high dimension space ,what’s the distribution of Chinese characters ? The paper discussed this problem through visualization. At the beginning ,we gave a definition of the Chinese characters’features that are used in this paper. Then by using K-L Transformation ,Linear Projection and Nonlinear Projection ,we made a visualization analysis to the distribution of some typical samples of Chinese characters in their feature space. The results showed that visualization analysis could help us understand the distribution of Chinese characters in their feature space ,thus being instructional to improving the performance of classifier.
-
Review2000, 14(5): 49-56.Text browsing is the assistant reading mechanism to help users browse the online texts. Text browsing based on Latent Semantic Indexing (LSI) is presented in this paper ,and it combines LSI with concept tagging to improve the efficiency of users reading. It applies LSI to reduce the skew intersections and calculates the similarity between terms and texts based on the semantic space ,it also divides the terms into several semantic classes and determines the meanings of classes. In additional ,it implements the information navigation based on conceptual tree.
-
Review2000, 14(5): 57-63.This paper discusses the reason that chinese text cannot be processed correctly in Java program from three aspects : character set ,Java complier and runtime enviroment . We provide a comprehensive solution for this question which is independent of all platforms ,including reading and writing files ,transfering chinese by network and inputting chinese.
