On Intelligent Making Chinese Character Based On Cognitive Mechenism

Information-based Character is an important part of the informationization. Current Chinese information processing method is mainly based on the character library. Chinese information processing system based on the character library takes Chinese characters as unit of information processing, but it has the following deficiency: (1) it is difficult to form an appropriate scale and long-term stability of the national and international standards, (2) it does not comply with Chinese making laws and it can not be defined a good heritage of Chinese civilization, (3) it does not meet Chinese cognitive laws, and it does not accord with content of existing Chinese teaching, (4) it can not fully meet the social application requirements.(5) It leads to high information entropy and low efficient of information processing.Intelligent Informationization is the characteristics of the 21th century. Cognitive science is disciplines to explore the essence of human intelligence.With development on intelligence science and cognitive science, some research methods and solutions will be come out to solve the current problem of Chinese information processing based on character library. Therefore, intelligent technology and results of research on cognitive science will be integrated to do an in-depth study for the theory and method on intelligent making Chinese character. The research work and the contribution of this paper are as follows.(1)Based on in-depth analysis for theory of traditional creating Chinese characters and cognitive mechanism of Chinese characters, theory of intelligent making Chinese based on cognitive mechanism is proposed through taking theory of creating Chinese characters and theory of prototype matching from cognitive psychology together. In the above study, based on a point set topology for the mathematical tools, the model of intelligent making Chinese character is built, and computer architecture of intelligent making Chinese character is designed.(2)Based on depth analysis for related study of the Chinese characters prototype, it is found that Chinese Mandarin and Chinese components are the main source of Chinese characters prototype. In the above study, principle of prototype extraction is proposed, which take into account the cultural heritage, ease of use and the efficiency of computer processing.According to the principle of prototype extraction, independent Chinese character and the Chinese Mandarin from "Ci Hai" is taken as the initial set of prototype, the experimental process of cycle optimization which accord to experiment, analysis and adjusting, is proposed. In the above experimental process, experiment about grouping Chinese characters with prototype is made for study of prototype extraction. In accordance with the above process, GB18030-2000 is taken as the target set to carry experiment of prototype extraction. A stage set of Chinese character prototype which cover all test set is obtained. The experimental results show that the extraction process of prototype is viable and has a good generalization, the extraction process of prototype can be extended to other prototype extraction of Chinese characters experimental set. To a certain extent, the feasibility of intelligent making Chinese character is identified by experiment of prototype extraction.(3)Based on the study of related research of Chinese characters structure ,aimed at deficiency of application to the current classification method for Chinese characters structure, classification theory of Chinese characters structure based on the graphic classification is proposed. In accordance with above classification theory of Chinese characters structure, with the original 13 kinds of Chinese characters structure, 18 kinds of classification method of Chinese character structure is proposed. GB2312-80 and GB18030-2000 are taken as experiment set for classifying Chinese structure. The feasibility of classifying method is verified by results of experiment, and classification method can cover all test set.From the point of view of computer processing, the mathematical model grid-based used to identify the Chinese character structure is established. With the above 18 kinds of Chinese classification method of Chinese character structure, GB2312-80 is taken as experimental set to carry the computer recognition experiment based on the mathematical model grid-based . Experimental results show that the proposed 18 kinds of Chinese classification method of Chinese character structure adapt to computer information processing.(4)Based on outlining the modern coding theory, with in-depth analysis of the Chinese cognitive mechanism and the traditional habit of Chinese characters used, research target of Chinese character coding is made. In the above study, Chinese character coding method for intelligent making Chinese character is proposed.The "Chinese characters structure + Chinese characters prototype" coding scheme is used by the coding method. And the coding method is in line with the cognitive structure of Chinese learners as follow, after the first partial overall, the first after the small, from top to bottom, from left to right, first, after the.According to the above coding method, GB18030-2000 is taken as the experimental set, with the computer-aided coding experimental platform , experimental processes with constantly optimize the adjustment is adapted to carry Chinese character coding experiment. The experimental results show that there are not repetitive Chinese character codes, the average length of code is less than 5, and Chinese character code contain information all the characteristics information of Chinese character. The experimental results also show that the desired research object is achieved. (5)According to the requirement of experiment for intelligent making Chinese character, the system requirement of intelligent making Chinese character experimental platform is analyzed, and technology schema of platform is given. Based on above schema, intelligent making Chinese character experimental platform is implemented on the Linux operating system.