[1]贾 畅,叶 飞,刘帅君,等.基于字向量和增强表示 BiLSTM 句子相似度研究[J].计算机技术与发展,2020,30(10):97-100.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 018]
　JIA Chang,YE Fei,LIU Shuai-jun,et al.Research on Sentence Similarity Based on Character Vector and Enhanced Representation BiLSTM[J].,2020,30(10):97-100.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 018]
点击复制

基于字向量和增强表示 BiLSTM 句子相似度研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
30
期数:
2020年10期
页码:
97-100
栏目:
智能、算法、系统工程
出版日期:
2020-10-10

文章信息/Info

Title:
Research on Sentence Similarity Based on Character Vector and Enhanced Representation BiLSTM
文章编号:
1673-629X(2020)10-0097-04
作者:
贾 畅; 叶 飞; 刘帅君; 麻之润
云南农业大学 大数据学院,云南 昆明 650201
Author(s):
JIA Chang; YE Fei; LIU Shuai-jun; MA Zhi-run
School of Big Data,Yunnan Agricultural University,Kunming 650201,China
关键词:
智能客服; 句子相似度; 循环神经网络; 字向量; 句子对齐
Keywords:
intelligent customer service; sentence similarity; recurrent neural network; character vector; sentence alignment
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2020. 10. 018
摘要:
目前分词工具在金融领域智能客服中无法对金融相关词汇进行有效切分,且基于单词的模型更容易受到数据稀疏性和词汇表外单词的影响。 针对该问题,提出一种基于字向量和增强表示 BiLSTM 的句子相似度计算模型—EBiLSTM。该模型首先通过双向长短时记忆网络 BiLSTM 提取由字嵌入组成的句子的字特征及其上下文表示, 然后计算句子对中一个句子与另一个句子的软对齐表示,在此基础上通过句子表示与其对齐表示间的交互来增强最终的句子表示。 所提模型可以有效学习到句子对的语义关系,加入增强表示层后通过两个句子的交互可以更好地捕捉两个句子间的语义差异。 实验表明,所提模型在真实数据集上,精确率、召回率和 F1 值均优于基于词向量的 CNN 和 BiLSTM 方法,也优于基于字向量的 CNN 和 BiLSTM 方法。
Abstract:
Currently word segmentation tools cannot effectively segment financial-related vocabulary in intelligent customer service in the financial field, and word-based models are more susceptible to data sparsity and out-of-vocabulary words. Aiming at this problem,EBiLSTM,a sentence similarity calculation model based on BiLSTM based on character vector and enhanced representation,is proposed. The model first extracts the word features and contextual representation of a sentence composed of words through a bi-directional longterm short-term memory network BiLSTM, and then calculates the soft-aligned representation of one sentence and another sentence in the sentence pair, and then aligns it with the sentence represen-tation. Inter-representation interactions enhance the final sentence representation. The proposed model can effectively learn the semantic relationship of sentence pairs. After adding the enhanced presentation layer,the semantic differences between two sentences can be better captured through the interaction of the two sentences. Experiment shows that the proposed model is better than the CNN and BiLSTM methods based on word vectors and the CNN and BiLSTM methods based on character vectors in terms of precision,recall and F1.

相似文献/References:

[1]王慧芝 安玉朋.基于常见问题库的多搜索引擎自动问答系统[J].计算机技术与发展,2006,(11):136.
　WANG Hui-zhi,AN Yu-peng.Multiple Search - Engine Question - Answering System With Frequently - Asked Question[J].,2006,(10):136.
[2]郑诚 夏青松 孙昌年.一种基于成分的句子相似度计算[J].计算机技术与发展,2012,(12):101.
　ZHENG Cheng,XIA Qing-song,SUN Chang-nian.Sentence Similarity Calculation Based on Composition[J].,2012,(10):101.
[3]李春梅,徐庆生.基于多特征的汉语句子相似度计算模型的研究[J].计算机技术与发展,2014,24(06):136.
　LI Chun-mei,XU Qing-sheng.Research on Chinese Sentence Similarity Calculation Model Based on Multi-features[J].,2014,24(10):136.
[4]闫红[],李付学[],周云[]. 基于HowNet句子相似度的计算[J].计算机技术与发展,2015,25(11):53.
　YAN Hong[],LI Fu-xue[],ZHOU Yun[]. Calculation of Sentence Similarity Based on HowNet[J].,2015,25(10):53.
[5]向 倩.基于聚类和相似度计算的陆空通话词向量评估[J].计算机技术与发展,2020,30(09):137.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 025]
　XIANG Qian.Word Embeddings Evaluation Based on Clustering and Similarity Computing in Radiotelephony Communications[J].,2020,30(10):137.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 025]
[6]杜睿山,张轶楠,田 枫,等.基于知识图谱的智能问答系统研究[J].计算机技术与发展,2021,31(11):189.[doi:10. 3969 / j. issn. 1673-629X. 2021. 11. 031]
　DU Rui-shan,ZHANG Yi-nan,TIAN Feng,et al.Research on Intelligent Question Answering System Based on Knowledge Map[J].,2021,31(10):189.[doi:10. 3969 / j. issn. 1673-629X. 2021. 11. 031]
[7]杨立鹏,廉文彬,季续国,等.12306 在线咨询服务智能应答研究[J].计算机技术与发展,2022,32(07):149.[doi:10. 3969 / j. issn. 1673-629X. 2022. 07. 026]
　YANG Li-peng,LIAN Wen-bin,JI Xu-guo,et al.Research on Intelligent Response of 12306 Online Consulting Service[J].,2022,32(10):149.[doi:10. 3969 / j. issn. 1673-629X. 2022. 07. 026]

更新日期/Last Update: 2020-10-10