《情态动词will语义排歧支持向量机人工神经网络特征提取》由会员分享，可在线阅读，更多相关《情态动词will语义排歧支持向量机人工神经网络特征提取（5页珍藏版）》请在金锄头文库上搜索。
1、 情态动词will论文：基于支持向量机的英语情态动词Will语义排歧研究【中文摘要】语义排歧是指根据目标词出现的上下文语境自动识别其意义。语义排歧是机器翻译、信息检索、语音识别、文本分类以及人机交互等诸多领域中的关键环节,是自然语言处理领域的热点和难点。尽管语义排歧技术取得了很大发展,但目前的语义排歧研究对象还是主要集中在普通动词和名词的语义排岐上。情态表达说话人的态度和意见,主要由情态动词来实现。因此,正确识别情态动词的语义对理解和领会说话人的态度和意见十分重要。情态动词语义有三种不确定性:梯度,歧义和融合。这些不确定性使人们很难把握其准确意义。因此,建立一个有效的、准确性较高的情态动词语义
2、排歧模型变得至关重要。本研究基于120万字的语料库,从will的实际使用语境中提取八个语义特征和句法特征,并采用数据挖掘中的一种新方法支持向量机,建立了情态动词will的语义排歧模型。实验结果显示,由支持向量机方法建立的情态动词will的语义排歧模型的排歧精度达到了98.33%。这个结果,证实了采用支持向量机对情态动词will语义排歧的有效性,同时证明了从真实的语料库中提取的8个语言特征的有效性。为了验证支持向量机语义排歧效果的优越性,本文采用神经网络技术中的反向传播神经网络,径向基神经网络.【英文摘要】Word sense disambiguation is the task to iden
3、tify the intended meaning of an ambiguous word in a certain context. Due to its wide application in machine translation, information retrieval, speech recognition, text categorization, it has been one of the hot and tough issues in natural language processing. Although techniques of word sense disam
4、biguation have advanced greatly, the research objects have mainly centered on the common nouns and verbs. Modality is concerned with the speakers opinion or attitude .【关键词】情态动词will 语义排歧 支持向量机 人工神经网络 特征提取【英文关键词】modal verb will word sense disambiguation Support Vector Machines artificial neural networ
5、k feature selection【索购全文】联系Q1：138113721 Q2：139938848【目录】基于支持向量机的英语情态动词Will语义排歧研究摘要5-7Abstract7-8Abbreviations12-13List of Tables13-14List of Figures14-15Chapter 1 Introduction15-191.1 Background of the present study15-171.2 Objectives of the present study171.3 Outline of the thesis17-19Chapter 2 Lit
6、erature Review19-372.1 Studies on the word sense disambiguation19-252.1.1 Studies on the word sense disambiguation abroad19-222.1.2 Studies on the word sense disambiguation in China22-252.2 Studies on the application of support vector machines25-312.2.1 Study on the application of support vector mac
7、hines abroad26-292.2.2 Study on the application of support vector machines in China29-312.3 Studies on the English modality31-352.4 Space for the present study35-37Chapter 3 Theoretical Foundation and Methodology37-413.1 Theoretical foundation of the present study37-383.2 Research method and data co
8、llection38-393.3 Summary39-41Chapter 4 Semantic Categorization of the English Modal Verb Will41-464.1 Why is will41-424.2 Categorization of meanings of English modal verb will42-444.3 Summary44-46Chapter 5 The Building of the WSD Model of Will by SVM46-585.1 The working principle of support vector m
9、achines46-485.2 Selection of training samples and test samples48-495.3 Construction of feature sets49-515.4 Vectorization of the linguistic features51-525.5 The building of WSD model by SVM52-575.5.1 Training and testing with default parameters52-545.5.2 Training and testing with optimal parameters5
10、4-565.5.3 Model selection56-575.6 Summary57-58Chapter 6 Comparative Analysis of the Models of WSD Models by SVM and by ANN58-736.1 The working principle of artificial neural networks58-606.2 The building of the WSD model by BP, RBF and PNN60-656.3 Comparative analysis of the four WSD models of will6
11、5-686.4 Analysis and discussion on the misclassified samples68-726.5 Summary72-73Chapter 7 Contributions of Different Linguistic Features to the WSD of Will73-887.1 The contributions of the semantic features to the WSD of will75-807.2 The contributions of syntactic features to WSD of will80-827.3 Th
12、e contribution of each linguistic feature to the WSD of will82-847.4 Validation of the importance of linguistic features to the WSD of will84-867.5 Summary86-88Chapter 8 Conclusions88-91References91-103Appendix I103-104Appendix II104-107Appendix III107-108Appendix IV108-112Appendix V112-113Appendix VI113-116Appendix VII116-119Appendix VIII119-122Acknowledgements122-123作者简介123
