Multi-language Instruction For Writing On The Integration Of Knowledge-building

Knowledge is the necessary resources of automatic text writing on teaching system. Currently, single languages repository technology is quite mature. Today’s society is more diversified and merging cultures is closer. In order to meet this requirement, writing on automatic teaching system should have the characteristics of multi-language compatible. Knowledge building multi-language compatible be come a research hotspot. Computer-implemented multi-language knowledge base to build and manage to overcome the difficulties:(1) the large number of multi-language text, to achieve more complex coding compression;(2) integration of the machine to achieve multi-language knowledge generated automatically;(3) multi-language text strokes, strokes relations, relations of both components have different characteristics in common. This paper discusses the integration of multi-language knowledge base construction, obtain the following results and innovative content:1、 For the common multi-language text strokes, strokes relations and relations of components and features, the strokes, stroke relationship with parts shared with the unique relationship into two parts, to distinguish processing, the problem is more complex disorder clarity, greatly reducing the amount of data encoding.2、 The proposed method is based on coding sequence. Taking into account the character strokes, strokes relations and relations component scalability, increased coding margin. In a shared basis with the unique integration of a complete coding margin coding system, fully consider and address the multi-language volume, complexity, scalability and other issues.3、 To achieve dominant stroke, cross-Stroke Relationship wrong, wrong parts from strokes relations and relations automatically generated. Current abbreviations and shortened stroke extend space, make the wrong prediction that may arise from the cross-strokes and stroke order wrong even right, to achieve coding. Two adjacent space on the stroke of any relationship does not exist, the use of parts separators, conducted after the whole word recognition component to finish, to give the component separators relationship coding.4、 The use of relation database for knowledge management. Proposed hierarchical database model based on knowledge of data characteristics. For writing on teaching knowledge entry point data can be read knowledge, knowledge modify the basic data manipulation capabilities.Experiments and analysis show that the calculated element coding space has autogenous mechanism automatically generated template text encoding computing elements and high reliability. Correct identification rate of strokes and all kinds of relationships have been greatly improved, the rate of entry and manual coding relatively efficient.