泰语句子相似度计算研究

发布时间：2018-05-21 17:51

本文选题：向量模型 + 句子语义　；参考：《昆明理工大学》2017年硕士论文

【摘要】：泰语句子相似度计算是信息处理领域的问题之一,随着国家“一带一路”战略的实施,我国与东南亚国家经济交往越来越密切,而泰国是东南亚地区重要国家之一,因此基于泰语的自然语言处理研究非常重要。泰语句子相似度计算在多个领域有广泛的应用,特别是泰语机器翻译以及专家系统领域,相似度计算是不可或缺的部分。在泰语句子相似度计算方法的研究过程中,本文从泰语句子的关键词、知网的语义、WordNet语料库的语义三个方面的特征进行了研究,针对以上的三个方面,本文提出了一种融合多特征的泰语句子相似度计算方法。最后,我们设计和实现了一个泰语句子相似度计算的应用系统,本文完成的主要工作有以下几个方面:1.深入分析了泰语句子相似度计算以及向量空间模型,设计了泰语分词的方法,并且利用此方法实现了泰语句子相似度计算的系统。2.研究了泰语词汇语义相似度及其计算方法。句子由词汇结合特定的语法规则构成,因此词语的相似度计算技术是研究句子相似度计算的基础,因此研究词语的相似度计算具有非常重要的意义。3.对现有的文本相似度计算技术进行了研究和分析,其中重点研究WordNet为语义知识库着重计算机对泰文句子的语义分析技术。4.设计实现了一个基于泰语句子相似度计算的系统,并用该系统设计实验验证了本文提出的泰语句子相似度计算方法的效率。
[Abstract]:The calculation of Thai sentence similarity is one of the problems in the field of information processing. With the implementation of the national "Belt and Road" strategy, the economic exchanges between China and Southeast Asian countries are getting closer and closer, and Thailand is one of the important countries in Southeast Asia. Therefore, the study of natural language processing based on Thai is very important. The calculation of Thai sentence similarity is widely used in many fields, especially in the field of Thai machine translation and expert system. In the course of studying the method of calculating the similarity of Thai sentences, this paper studies the features of the key words of Thai sentences and the semantic features of WordNet corpus, aiming at the above three aspects. In this paper, a method for calculating the similarity of Thai sentences with multiple features is proposed. Finally, we design and implement a Thai sentence similarity calculation application system. The main work of this paper is as follows: 1. This paper deeply analyzes Thai sentence similarity calculation and vector space model, designs the method of Thai word segmentation, and realizes the system of calculating Thai sentence similarity by using this method. The semantic similarity of Thai vocabulary and its calculation method are studied. Sentences are composed of words combined with specific grammatical rules, so the technology of word similarity calculation is the basis of sentence similarity calculation, so the study of word similarity calculation is of great significance. The existing text similarity computing techniques are studied and analyzed, in which WordNet is the semantic knowledge base and the computer semantic analysis technology for Thai sentences. A system based on Thai sentence similarity calculation is designed and implemented, and the efficiency of the proposed method is verified by the design of the system.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.1

【参考文献】