主要作者:宋双永 阿里达摩院 算法专家
编辑整理:Hoh
内容来源:作者授权
出品平台:DataFunTalk
注:欢迎转载,转载请留言。
01
智能客服系统中情感分析技术架构
(a)
(b)
图1:智能客服系统中的情感分析技术架构
02
用户情感检测
图4:SWEM模型中的最大值池化和平均值池化
图5:基于卷积的二元、三元和四元语义特征抽取
图6:LEAM模型中词语和类别标签之间的语义交互(传统方法和LEAM模型的对比)
表1:集成模型和三种baseline模型的效果对比
03
用户情感安抚
图7:用户情绪安抚整体框架
图8:用户情绪安抚示例
图9:文本匹配模型
图10:基于检索的问答系统流程图
表3:主题分类效果对比
表4:负面情绪安抚对用户满意度的效果对比
表5:感激情感安抚对用户满意度的效果对比
04
情感生成式语聊
图12:智能客服系统中的情感生成式语聊模型
图13:生成式语聊中的注意力机制
图14:小蜜空间中的情绪生成式语聊应用实例
05
客服服务质检
图15:客服服务质检模型1
表7:质检模型两种上下文语义抽取方式对比
06
会话满意度预估
图17:智能客服系统中的用户会话满意度预估模型
图18:用户会话满意度预估结果比较
表9:用户会话满意度预估结果比较
07
智能人工入口
图19:智能客服系统中的智能人工入口用户排序模块
图20:基于DeepFM的DialogueEncoder
表10:使用会话特征
08
总结与展望
今天的分享就到这里,谢谢大家。
如果您喜欢本文,欢迎点击右上角,把文章分享到朋友圈~~
社群推荐:
欢迎加入 DataFunTalk NLP 交流群,跟同行零距离交流。如想进群,请加逃课儿同学的微信 ( 微信号:datafun-coco ),逃课儿会自动拉你进群。
宋双永
阿里巴巴 | 达摩院算法专家
shuangyong.ssy@alibaba-inc.com
王超,阿里花名王栩,资深算法工程师
陈成龙,阿里花名谌龙,算法专家
参考文献:
[1] Okuda T, Shoda S. AI-based Chatbot Service for Financial Industry [J]. Fujitsu Scientific and Technical Journal, 2018, 54(2): 4-8.
[2] Zhu X. Case II (Part A): JIMI’s Growth Path: Artificial Intelligence Has Redefined the Customer Service of JD.com [M]. In Emerging Champions in the Digital Economy. Springer, Singapore, 2019: 91-103.
[3] Li F-L, Qiu M, Chen H, et al. AliMe assist: an intelligent assistant for creating an innovative e-commerce experience [C]. In Proceedings of the ACM on Conference on Information and Knowledge Management, 2017: 2495-2498.
[4] Shen D, Wang G, Wang W, et al. Baseline needs more love: On simple word-embedding-based models and associated pooling mechanisms [C]. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 2018: 440-450.
[5] Zeng D, Liu K, Chen, Y., et al. (2015). Distant supervision for relation extraction via piecewise convolutional neural networks [C]. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, 2015: 1753-1762.
[6] Wang G, Li C, Wang W, et al. Joint Embedding of Words and Labels for Text Classification [C]. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 2018: 2321–2331.
[7] Pang L, Lan Y, Guo J, et al. Text Matching as Image Recognition [C]. In Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, 2016: 2793-2799.
[8] Yin W, Schütze H, Xiang B, et al. Abcnn: Attention-based convolutional neural network for modeling sentence pairs [J]. Transactions of the Association for Computational Linguistics, 2016(4): 259-272.
[9] Young T, Cambria E, Chaturvedi I, et al. Augmenting end-to-end dialogue systems with commonsense knowledge [C]. In Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018: 4970-4977.
[10] Bahdanau D, Cho K, Bengio Y. Neural machine translation by jointly learning to align and translate [C]. In 3rd International Conference on Learning Representations, 2015: .
[11] Zaremba W, Sutskever I, Vinyals O. Recurrent neural network regularization [C]. In 3rd International Conference on Learning Representations, 2015.
[12] 庞亮, 兰艳艳, 徐君,等. 深度文本匹配综述[J]. 计算机学报, 2017, 40(4):985-1003.
[13] Hu B, Lu Z, Li H, et al. Convolutional neural network architectures for matching natural language sentences [C]. In Advances in neural information processing systems, 2014: 2042-2050.
[14] Lu Z, Li H. A deep architecture for matching short texts [C]. In Advances in Neural Information Processing Systems, 2013: 1367-1375.
[15] Qiu X, Huang X. Convolutional neural tensor network architecture for community-based question answering [C]. In Twenty-Fourth International Joint Conference on Artificial Intelligence, 2015: 1305-1311.
[16] McCandless M, Hatcher E, Gospodnetic O. Lucene in action: covers Apache Lucene 3.0 [M]. Manning Publications Co.2010.
[17] Guo H, Tang R, Yey Y, et al. DeepFM: A Factorization-Machine based Neural Network for CTR Prediction [C]. In 26th International Joint Conference on Artificial Intelligence, 2017: 1725-1731.
[18] Chen C, tensorflow-XNN [EB]. 2018. https://git hub.com/ChenglongChen/tensorflow-XNN.
[19] Song S, Meng Y, Shi Z, et al. A simple yet effective method for summarizing microblogging users with their representative tweets [C]. In 2017 International Conference on Asian Language Processing, 2017: 310-313.
[20] Yu J, Qiu M, Jiang J, et al. Modelling domain relationships for transfer learning on retrieval-based question answering systems in e-commerce [C]. In Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining, 2018: 682-690.
[21] Zhou H, Huang M, Zhang T, et al. Emotional chatting machine: Emotional conversation generation with internal and external memory [C]. In Thirty-Second AAAI Conference on Artificial Intelligence, 2018: 730-738.
[22] Li J, Galley M, Brockett C, et al. A personabased neural conversation model [C]. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2016: 994–1003.
[23] Wang Y, Huang, M, Zhao L. Attention-based LSTM for aspect-level sentiment classification [C]. In Proceedings of the 2016 conference on empirical methods in natural language processing, 2016: 606-615.
[24] Zhou L, Gao J, Li D, et al. The Design and Implementation of XiaoIce, an Empathetic Social Chatbot [R], 2018, arXiv preprint arXiv:1812. 08989.
情感分析算法在阿里小蜜的应用实践
关于我们:
一个在看,一段时光!👇