The Construction Of Mongolian Chinese English Language Knowledge Base

The paper constructed a Mongolian Chinese English language knowledge base. Paper mainly introduced the "Mongolian Chinese English Dictionary" database, the Lexical variation in Mongolian library, construction of multilingual aligned corpus and database platforms and the invention of user interface of Mongolian Chinese English Knowledge Base and so on."Mongolian Chinese English Dictionary" contains total of 20,551 records, set up 25 attribute fields such as form of word, phonetic symbols, pronunciation, structure, parts of speech, semantic class of word, the target language, connected digital of connection with other data base and so on."Mongolian morphological forms database" contains total of 2,442,960 records. Recorded Mongolian morphological forms of each words of "Mongolian Chinese English Dictionary". It set up 10 attribute fields such as serial number, root, root of speech, lexical variation and so on. The database records are automatically generated by rules and programming.Multilingual aligned corpus was constructed in two forms, namely as a text file libraries and Access 2007 database. It contains a total of 4174 pairs of one-directional alignment Mongolian Chinese sentences and 25,000 of unidirectional alignment Mongolian Chinese English sentences. These sentences were selected from contemporary formal publications. In Access 2007 database in the form of building a corpus, were set up number, Mongolian sentences, Chinese sentences, English sentences, the source publications of 9 attribute fields.There are three purposes for Mongolian Chinese English language knowledge base construction:First, provides tools for the Mongolian mother tongue and second language teaching and provides resources for second language teaching such as Chinese and English. Second, provides knowledge of language for cross-language retrieval of Mongolian language. Third, provides translation dictionaries, and aligned corpus for the machine translation system. For these purposes, the same expression object set a few properties field. For example phonetic symbols were set up two attribute fields, namely the Latin transliteration form formulated by Inner Mongolia University and the Latin transliteration form used by "Mongolian orthography dictionary (Revised edition)".Building database platform consists of speech tagging, tagging compound word category, generation of lexical variation, connected digital of connection with the corpus, connected digital of connected with library and lexical variation corpus, word form generation, building corpus indexing function modules which are used for building the database, so it is named as a building base platform.User interface of Mongolian Chinese English language knowledge base was constructed for providing ease for user. It has the function of exact queries and fuzzy queries. The query word can stand Menksoft input method, you can also use Microsoft Mongolian input method, or with a Latin transliteration form of Inner Mongolia University. The display items contain Cyril Mongolian, Chinese, English, homograph, homonym, the same root word, the same semantic category words, grammatical changes, sentences, terminology, etc. Can show all, or one or more may be displayed.The paper also described the problems and solutions encountered when building a database.