Interweaving of Syntax and Semantics in Algorithms For Recognising Chinese Characters David Al-Dabass 1 And Manling Ren 2 1 Dept of Computing, The Nottingham Trent University, Nottingham, NG1 4BU, UK david.al-dabass@ntu.ac.uk 2 Xerox Limited Technical Centre, Welwyn Garden City, Hertfordshire AL7 1HE, UK manlingren@hotmail.com Abstract: The structure of Chinese characters is reviewed and seen to be best represented as 3-layer hierarchy of character, radical and stroke. Fuzzy possibilistic reasoning is then put forward as an appropriate set of conceptual tools to investigate the automatic recognition of these characters. An associative memory artificial neural network algorithms form a suitable technique for realising these concepts. Implementing these techniques several issues are explored: vagueness of radicals, their situation, position invariance, extraction order and shape. Extensive results are obtained to demonstrate the quality of the algorithms in dealing with the range of difficulties inherent in the problem. Keywords: Chinese character recognition, fuzzy possibilistic reasoning, associate memory neural network, topological structure. 1 Introduction The complex structure of Chinese characters is formulated through a long history (about 5,500 years recorded in history). Early Chinese characters were mainly symbols and pictographs that could also represent some abstract concepts of daily life as shown in Figure 1 (a). In order to express more complex ideas and concepts, pictographs were developed and combined to form ideographs for multiple meanings. These ideographs form some 90% of the total Chinese characters in current usage [Scu91]. Most ideographs are made up of two components: (a) a radical, i.e. a pictograph before it becomes part of an ideograph, which indicates the classification of a character; and (b) a ‘phonetic’ symbol for partially aid the pronunciation of a character. Figure 1 (b) shows several examples of the ideographs development. Chinese characters possess three major features in their structures and quantities: a two- dimension (2-D) pictorial format, topological structure and a large vocabulary. Figure 2. 2-D pictorial format of Chinese characters Figure 1. The historical development of Chinese characters Figure 3. Topological structure of Chinese characters