Font Size: a A A

The Semantic Relation Pattern Of "V[Double Syllables]+V[Double Syllables]" & Automatic Recognition

Posted on:2005-04-14Degree:MasterType:Thesis
Country:ChinaCandidate:F L YanFull Text:PDF
GTID:2155360122498291Subject:Chinese Philology
Abstract/Summary:PDF Full Text Request
This thesis presents the problem about the semantic relation pattern of two V[double syllables] in modern Chinese, reviews the specific semantic relation types and attempts to put forward the arithmetic that can automatic recognize the semantic relation pattern of "V[double syllables]+ V[double syllables]".Corpus are important resources for knowledge acquisition in the field of natural language processing. This thesis makes use of a large-size corpus that is at a scale of 12 million words from uses the sequential verbs templates from for reference, induces 8 kinds of semantic relation, which are compound, parataxis, object, content, result & affair, patient, sequence, mood, etc.The highlight of this thesis is its ability to effectively describe various relations between Chinese words. All of these profited from using for reference and the combination with specific use of language. The thesis gives an effective account on the semantic relation of "V[double syllables]+ Vfdouble syllables]" in the running linguistic environment. And it uses repository and the sequential verbs templates from as the base of semantic calculation, designs an automatic recognition arithmetic. The result of experiments indicates that this approach is feasible.The results of this thesis can be applied in many aspects, such as machine translation, linguistic analysis, etc. Based on the arithmetic as described in this thesis, a system is designed. The system is programmed in Delphi 6.0 and its database is designed in Paradox 7. 0. And this thesis also gives an evaluation of the arithmetic.
Keywords/Search Tags:"V[double syllables]+V[double syllables]", automatic recognition of the semantic, relation pattern, Chinese information processing, corpus HowNet, Machine, Translation similarity, pattern matching
PDF Full Text Request
Related items