The Representation Of Chinese Semantic Knowledge And Its Application In The Chinese-English MT System

Inthls thesis,we researchthe semm九Ic level ofnatuxal language processing．"乍propose a hierarchy to represent Chinese sem四万Ic仙owledge,design a method toacquire ltanduse It Inthe Interlingua－basedChlneseto Engllshmachlnetranslatlonsystem called ICENTll． Semantic analysis lsthebasls ofnatural language understanding．他 discussdi肚rent sem山uIc theories and劝owledge representation methods,and then we proposeand construct a frame－based Chinese semantic knowledge hierarchy．The hierarchy Iscomposed ofsem咖Ic units and sem咖Ic rules．The sem咖Ic unit Is composedofsemantic prlmltlves,senses and semantic chunks．The semantic prlmltlve Is the basicunit of the hierarchy and describes all kinds of semantic features．The sense expressesconcepts and Is descrlbedbyprlmltlves．The semantic chu尬 expresses compoundconcepts and Is decrlbedbynested fraxnes．In semantic rames,sem咖Ic relations ofeach componemare expressed by sem咖Ic roles．Semantic rules are the劝stract ofsemttlchowledge and are composedofwell－formed sem出方Ic 11讪s and generatedchu业 templets．Wei－formed semantic 11讪s descnbe the semantic restriction amongsemantic units．Generated oh帆 templets describe the semantic structure which Isgenerated by well－formed semanlc links．Based on the semanlc hierarchy we designand lmpleme血 an algorlthln ofsem8lltlc ule acquiring,which acquires well－formedsemantic 11冰s fromthe combination Instance corpus 用nomatlcally and acquiresgeneratedchunktemplets semi－automatlcany The system ICENTll Is an Interlingua－basedMT system,whlchls lmplemefltdbyadding sem山iIcAnowledge Inthe preceding system ICENT The Interlingua design Isthe key Inthe design of＊the system．All results ofChlnese analysis are expressed Inthel血erlingua and all Information ofEngllsh generation can be got In It．We discussdlfferentphenomenabetween Chinese andEngllsh,design and lmplemeniasyntactlcally and semanilcally combined Interlingua．The血erlingua Is experessed byframe structures,which contains some slot－value pairs to describe syntactic andsemantic attributes．The Interlingua ofIC刚Tll emphasizes semantic represefltatlon,Which contains the sense ofa word and the semantic relations ofap价ase or a sentence．The Interlingua ofICENTll combines syntactic and sem咖Ic Information,conceptgranularltyls well situated,expression lspersplcuous and implementation Is simple．The illterlingua ofICmTll Is complemellted and IWroved In the process of＊thes designof＊the一一 system,and Is provedto be劝leto meetthe needs ofthetranslatlon． Segmentation and tagging Is the6rst step of1CENENTn．V厄 design and IInpkment ana川omatlc sem山上Ictagging algorulllTI to tag sense forthe resultofsegmeniatlon．hesemanlc tagging deals with monoseme, polyseme and unknown word respectively anduses sy血actlc and semantic他owledge to determine the sense tagging set for eachword．The resultcanbe non－single and sense dlsamblguatlon canbe done Inthe semanticanalysis．Ifthe sense tagging setconialnsthe correct sense,semantlctagginghltsthe 第＊页target. If the sense tagging set only contains the correct sense, semantic tagging is correct. The experiment is done and proves the algorithm can get a very high hit rate and a higher correctness rate. Usually, the size of hit sets will become rather small.The relation between Syntax and semantic meaning is very close. Syntax is the form and semantic meaning is the content. We combine syntactic parsing and semantic analysis, use the extended context free grammar to analyze Chinese. Each syntactic production has a precondition judge function. When the analyzer needs to use a production to do reducing, the precondition judge function corresponding to the production is activated and in it the corresponding semantic analysis is performed. The reduction can be done only if the semantic analysis is reasonable, otherwise the current analysis tree should be termin...