简单语义单元的语义修饰关系图模型及其求解
Graph Model of the Descriptive Semantic Relationship for the Simple Semantic Units and Its Solution

DOI: 10.12677/CSA.2020.103042, PP. 408-417

Keywords: 语义相关度，简单语义单元，语义修饰关系图，最大生成树变量
Semantic Relevancy, Simple Semantic Units, Graph of the Descriptive Semantic Relations, The Variable of the Maximum Spanning Tree

Full-Text   Cite this paper   Add to My Lib

Abstract:

为了有效地使用语义信息来进行自然语言处理，提出了一个简单语义单元的语义修饰关系图模型并对其求解方法进行了研究；首先提出了简单语义单元的概念，分析了简单语义单元的特点；在生成其语义修饰关系图的基础上，通过近似算法，求取语义修饰关系完全图的最大生成树；从而实现对简单语义单元的语法分析和语义消歧。实验表明，该方法具有一定的可行性。
In order to use semantics more effectively in natural language processing, a graph model of the descriptive semantic relationship for the simple semantic units was proposed and its solution was researched. Firstly the definition of the simple semantic unit was given and its characteristics were discussed. Then the graph of descriptive semantic relations could be created for a simple semantic unit, and the maximum spanning tree of the graph could be got by an approximate algorithm. And the simple semantic unit could be parsed and the word sense could be disambiguated. Finally experiments were finished and the results suggested that the method might be feasible.

References

[1]	陈谦. 基于神经网络的自然语言语义表征方法研究[D]: [博士学位论文].合肥: 中国科学技术大学, 2018.
[2]	彭敏, 杨绍雄, 朱佳晖. 基于双向LSTM语义强化的主题建模[J]. 中文信息学报, 2018, 32(4): 40-49.
[3]	贾玉祥, 许鸿飞, 昝红英. 基于神经网络的语义选择限制知识自动获取[J]. 中文信息学报, 2017, 31(1): 155-161.
[4]	Mikolov, T., Chen, K., Corrado, G., et al. (2013) Efficient Estimation of Word Representations in Vector Space. http://cn.bing.com/academic/profile?id=6b535fdb1fad7c57b8421d3a81b00af5&encoded=0&v=paper_preview&mkt=zh-cn
[5]	Wan, S., Lan, Y., Guo, J., et al. (2016) A Deep Architecture for Semantic Matching with Multiple Positional Sentence Representations. https://www.researchgate.net/publication/285271115_A_Deep_Architecture_for_Semantic _Matching_with_Multiple_Positional_Sentence_Representations
[6]	Pennington, J., Socher, R. and Manning, C. (2014) Glove: Global Vectors for Word Representation. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, Doha, Qatar, 25-29 October 2014, 1532-1543. https://doi.org/10.3115/v1/D14-1162
[7]	孟凡擎, 鹿文鹏, 张旭, 等. 基于HowNet的图模型词义消歧方法[J]. 齐鲁工业大学学报, 2018, 32(6): 69-76.
[8]	唐善成, 马付玉, 张镤月, 等. 采用Seq2Seq模型的非受限词义消歧方法[J]. 西北大学学报(自然科学版), 2019, 49(3): 351-355.
[9]	夏小强, 邵堃. 基于语义关系约束和词语关系信息的句向量研究[J]. 计算机应用研究, 2019, 36(7): 2023-2026.
[10]	Liu, Y.T. and Sun, H. (2015) Word Sense Disambiguation for Chinese Based on Semantics Calculation. Mathematical Problems in Engineering, 2015, Article ID: 235096. https://doi.org/10.1155/2015/235096

Full-Text