1999 Volume 13 Issue 1
Published: 15 February 1999
-
Select all|
-
Review1999, 13(1): 2-9.There are too many syntactic ambiguities in the Chinese , and it is not easy to parse the Chinese sentences automatically only with the Part of Speech (POS) tagging. In order to overcome this barrier we looked for the possible syntactic relations between the semantic classes , then built a Chinese Semantic Associate Net (CSAN) . Using the CSAN we find a way to solve the problem about the syntactic ambiguities. In this way we do not try to solve all the ambiguities , we only try to solve the certain kind of ambiguities using a special method. So our method is effective.
-
Review1999, 13(1): 10-16.Two methods for text classification based on fuzzy techniques are presented in the paper. A concept of‘fuzzy classification network’is used in the text classification. To meet the needs of these two classification methods ,an approach based on statistics and natural language understanding for automatically extracting the keywords from the text is presented. It is point out that the methods are also suitable for pattern recognition and text retrieval.
-
Review1999, 13(1): 17-20.In this paper , six principles of desiging syntax rules base of Russian - Chinese Intelligent machine translation system are proposed : 1 、predicate verbs as the centre of syntax structure ; 2 、complatible use of syntactic and semantic information ; 3 、give preference to usual sentence order ; 4 、express Russian syntax concord relations with functions ; 5 、forced unification of language forms ; 6 、Targer language is generated in the process of source language analysis.
-
Review1999, 13(1): 21-27.This paper introduces a Chinese glyph editor based on object - oriented technique. First it introduces the system model , classes and objects design. Then it makes a case for implementation of edit function , which is the communication mechanism of objects. Furthermore it compares the design method of traditional graphics and object - oriented technique.
-
Review1999, 13(1): 28-35.The solution of crossing ambiguities is still an open issue in the study of Chinese word segmentation. In this paper , we introduce the concept of maximal crossing ambiguity at first , divide it further into two major types , i. e. , the true and the pseudo. Having observed a Chinese corpus with 100M characters , we find that the high frequent part of maximal crossing ambiguities is strong in coverage capacity (the coverage of the top 4,619 is as high as 59.20% , out of which 4,279 belongs to the pseudo type , with coverage 53.35%) and rather stable with regard to domain shifting. As a consequence , we propose for high frequent maximal crossing ambiguities a memory - based strategy that is expected to improve the performance of practical Chinese word segmenters significantly.
-
Review1999, 13(1): 36-43.Computer - Assisted lnstruction (CAI) can eliminate the limitation of time and space.By the powerful ability of computers to process data rapidly and accurately ,it can bring the teaching and researching activities into full play. In this reason ,we can design and develop a lot of assistant teaching materiala and tools creatively for Chinese literature. For the students studying Chinese literature and those people that are interest in authoring poetry ,the biggest problem is the unfamiliarity with the catalogue of lyrics (唐宋词) ,the pattern of lyrics ,and the rhyme of lyrics.With the help of computer. we can develop a system with the rules of lyrics , the catalogue , the rhyme and pattern database , and the examples. After selecting the catalogue and the rhyme of lyrics by the users. this system can take over the task of the index of rules of lyrics and the rhyme of lyrics. When the users write the lyrics ,the system also can check the rules and find out the error in the works ,and prompt to user for modification. The system can also search for the examples and the works of Tong & Song lyrics ,and take them as reference. It gives the great help for the instruction of lyrics and practices. On the other hand ,it is an extremely valuable experience to implement the ideas for original creators and design a workable system to satisfy users in Chinese literature. We wish to make a good example of integrating the ideas and harmonizing the conflicts about the humanities and technology ,and to dedicate to our goal in the future.
-
Review1999, 13(1): 44-50.The analysis of amount property of Chinese character and phrase has became a main problem which should to be resolved immediately as ISO10646 has been played a more important role in the Chinese Information Processing. This paper mainly describes a model of the analysis and recombination about the property of Chinese character and phrase which based on Client/Server architecture , and also introduces the technique of the system design.
-
Review1999, 13(1): 51-59.For the existing problem in Basic sets and auxiliary sets builed on different plane and different system of the tibetan coded character sets ,We Present an extended tibetan coded character system - - Full Tibetan Coded Character System. In this paper ,we have also given the rule and method for composing and decomposing of the tibetan and ,implemented the information interchanging between the Basic sets and auxiliary sets by means of internal , code converting table. This new system can produce the normal and standard external tibetan word by means of component sets and ,it is cmpatible with UCS and GB2312. The author suggests building the extended Code Character system in the A - zone of the Basic multiligual plane.
-
Review1999, 13(1): 60-65.In Chinese numeral count system ,grammar analysis is a important part . In this paper ,a grammar analyzer of Chinese numeral is designed. It consists of the main cont rol part ,the knowledge part ,the manage part ,and the error warn part . The method recommended here consists of the following steps : 1. Get the BNF structure of Chinese numeral. 2. Get the syntax picture of Chinese numeral according to its BNF structure. 3. Check the syntax of input Chinese numeral and give the check answer : true or false.
