Title :
Mathematical-expressions-based knowledge discovery method and its application in verification of incompatibility of water and fire
Author :
Sun, Xingming ; Chen, Huo-Wang ; Yin, Jian-ping ; Jing, Xin-hai
Author_Institution :
Dept. of Comput. Sci. & Eng., Zhuzhou Inst. of Technol., Hunan, China
Abstract :
It is well known that water and fire are incompatible, which is described by a Chinese idiom. Because the Chinese character is an ideographic and pictographic character, we try to derive this law from our knowledge of the structure of Chinese characters. By analyzing the structure of Chinese characters deeply, we present a novel mathematical expression method of Chinese characters. Useful knowledge can be mined from the database of the mathematical expressions of all the Chinese characters. Our experimental results show that the probability of both the component ´water´ and the component ´fire´ appearing in a Chinese character simultaneously is very small. Even if the two components do appear simultaneously, we show from the figure of the character that ´water´ will boil and overflow if fire is burning. Thus, our experimental results give a reasonable explanation to the idiom ´water and fire are incompatible´.
Keywords :
character recognition; data mining; Chinese character structure; Chinese characters; knowledge mining; mathematical expression database; mathematical-expression-based knowledge discovery method; water fire incompatibility; Application software; Character recognition; Computer science; Cybernetics; Data structures; Databases; Fires; Machine learning; Prototypes; Software engineering;
Conference_Titel :
Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on
Print_ISBN :
0-7803-7508-4
DOI :
10.1109/ICMLC.2002.1174414