Description Of Attribute-Meaning And Construction Of Ontology Of The Mass Noun "Sunshine,Air,Water"

In this paper,the mass noun "sunshine,air,water" in Peking University CCL Corpus as corpus,the Generative Lexicon Theory as the theoretical basis,the Cognition-combinatory Meaning Outlook and Spherical Structure of Word Meaning as the direct basis,though comprehensive investigation of the original usage of the mass noun"sunshine,air,water".according to the specific use environment of words,analysis the cognitive motivation and semantic association mechanism,we eventually extract 276 kinds of specific attribute meanings of the mass noun "sunshine,air,water",including 112 kinds of specific attribute meanings of "sunshine",161 kinds of specific attribute meanings of "air",196 kinds of specific attribute meanings of "water".We further conclude these specific attribute meanings for 10 kinds of initial attribute meanings,and respectively construct the attribute meaning ontology of the mass noun "sunshine,air,water".This attribute meaning ontology not only completely present the actual usage of the mass noun "sunshine,air,water" in language,but also present the knowledge system that people how to know the three things.This paper is a preliminary practical result of attribute meaning ontology of the noun.On this basis,this paper further discuss the application values of attribute meaning ontology of words.This paper insists that the attribute meanings of words play the role of bridges when semantic structures extend and generate into syntactic structures,it can explain the cognitive semantic motivation when words are adjacent combined to produce dynamic portfolio significance.Meanwhile,this paper insists that the attribute meaning ontology establish a cognitive model which can help us more precisely and more comprehensively understand meanings of the noun,completing language modeling of syntactic and semantic knowledge of the noun.The attribute meaning ontology includes a multi-level semantic knowledge system which can provide foundational knowledge resources for natural language processing and application,such as information extraction,content analysis,text calculating and so on,also for solving the problems in syntactic-semantic interface.