Research On Semantic Multilingual Information Processing Platform-SMIPP

With the developing trends of world globalization and the increases in international cooperation and intercommunion, Nowadays the research on information processing focuses on the multilingual information, especially on its supported platform-Multilingual Information Processing Platform. At the same time, with the rapid developing of computer technology and the exponential growth in information size, the interests of character information processing also have been transferred from the simple characters input, output and storing to the content and meaning of the characters. Therefore, the research on the multilingual information processing platform which focused on content and semanteme not only has the practical values, but also has bright futures.A model of semantic multilingual information processing platform (shorted as SMIPP) oriented to information processing is put forward and this model provides not only a multilingual processing environment, but also a sorts of related techniques, such as encoding scheme, Ontology, corpora, input model, output model, etc.Firstly, to meet the request to express the semanteme of the characters for SMIPP, a multilingual encoding scheme named as SemaCode is put forward. SemaCode consists of seven layers, including physical storage layer, exchange and transmission layer, character code point layer, phrase code point layer, property layer, semantic layer and application layer. SemaCode is introduced a new encoding method for the design of character point layer. Following that method, language type has been encoding into the code point and each character (not glyph) has been assigned a code point. That method makes the SemaCode more adaptive to the information processing. On the propriety layer, the property tags have been applied to tag the characters, and consequently the SemaCode has the ability to mark the characters and...