[1]王乾铭,李 吟.基于深度学习的个性化聊天机器人研究[J].计算机技术与发展,2020,30(04):79-84.[doi:10. 3969 / j. issn. 1673-629X. 2020. 04. 015]
 WANG Qian-ming,LI Yin.Research on Personalized Chatbot Based on Deep Learning[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2020,30(04):79-84.[doi:10. 3969 / j. issn. 1673-629X. 2020. 04. 015]
点击复制

基于深度学习的个性化聊天机器人研究()
分享到:

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

卷:
30
期数:
2020年04期
页码:
79-84
栏目:
智能、算法、系统工程
出版日期:
2020-04-10

文章信息/Info

Title:
Research on Personalized Chatbot Based on Deep Learning
文章编号:
1673-629X(2020)04-0079-06
作者:
王乾铭李 吟
江苏自动化研究所,江苏 连云港 222006
Author(s):
WANG Qian-mingLI Yin
Jiangsu Automation Research Institute of CSIC,Lianyungang 222006,China
关键词:
聊天机器人Seq2Seq 模型注意力机制多样性个性化
Keywords:
chatbotSeq2Seq modelattention mechanismdiversitypersonality
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2020. 04. 015
摘要:
为了对传统的聊天机器人局限性进行改进,增强其回复时的个性化和多样化,并使其具有一定的准确性,研究改进了一种基于深度学习 Seq2Seq 模型的对话系统。 对传统的编码器-解码器(Encoder-Decoder) 模型进行了研究,在原有模型的基础上使用了深度 LSTM 结构并且加入注意力机制使其能更好地适应不同长度的问句。 在解码过程中,将原有模型的贪心算法改为了 Beam Search 算法。在训练过程中,利用了多次训练的方法,训练出模拟电视剧角色的聊天机器人,为聊天机器人赋予一个特定的身份。 通过使用两种自动评估指标 BLEUs 和 Distinct-n 去测试聊天机器人,并使用一些问句测试聊天机器人的反应,最终实验结果表明新模型与原有的模型相比,两种指标都取得了较好的效果,并且输出句子的合理性以及回复质量也有明显的提高。
Abstract:
In order to improve the limitations of traditional chatbots,enhance the personalization and diversity of their replies,and make them accurate,a conversation system based on deep learning Seq2Seq model is improved. The traditional Encoder-Decoder model is studied,and the depth LSTM structure is used on the basis of the original model and the attention mechanism is added to make it better adapt to the questions of different lengths. In the decoding process,the greedy algorithm of the original model is changed to the Beam Search algorithm. In the training process,multiple training method is used to train the characters in TV series and give chatbot a specific identity. Two automatic evaluation indexes, BLEUs and Distinct-n, are used to test the chatbot whose response is tested by somequestions. The final experiment shows that compared with the original model,the new model has achieved better results and improved the recovery quality.

相似文献/References:

[1]申静波,李井辉,孙丽娜.注意力机制在评论文本情感分析中的应用研究[J].计算机技术与发展,2020,30(07):169.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 036]
 SHEN Jing-bo,LI Jing-hui,SUN Li-na.Research on Application of Attention Mechanism in Comment Text Emotional Analysis[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2020,30(04):169.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 036]
[2]李 斌,王浩畅.智能对外汉语学习系统的设计与研究[J].计算机技术与发展,2022,32(03):15.[doi:10. 3969 / j. issn. 1673-629X. 2022. 03. 003]
 LI Bin,WANG Hao-chang.Design and Research of Intelligent Foreign Chinese Learning System[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2022,32(04):15.[doi:10. 3969 / j. issn. 1673-629X. 2022. 03. 003]
[3]李修贤,王桂玲,石永鹏.面向主动式 BPM 的 IoT 服务动态绑定方法[J].计算机技术与发展,2023,33(09):64.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 010]
 LI Xiu-xian,WANG Gui-ling,SHI Yong-peng.A Dynamic Binding Method for IoT Services towards Proactive BPM[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2023,33(04):64.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 010]

更新日期/Last Update: 2020-04-10