【原创】人工客服会话日志挖掘论文调研

业务需求:
  • 在人工客服服务日志中抽取问答对,配置到机器人知识库中
调研:
业界方案一:抽取式:
  • QA matching:以question为出发点,即假设question已经确定,从上下文(主要是上文)中找到该question的答案,组成问答对
    • 【ECAI-2020】Matching Questions and Answers in Dialogues from Online Forums:https://arxiv.org/pdf/2005.09276.pdf
    • 【LREC-2020】Cross-sentence Pre-trained Model for Interactive QA matching:https://www.aclweb.org/anthology/2020.lrec-1.666/
  • conversation structure modeling(dialogue structure analysis/conversation structure discovery/conversation disentanglement/discourse parsing):以answer为出发点,即假设answer已经确定,从上下文(主要是上文)中找到该answer对应的问题(reply-to relation),组成问答对,目前针对multi-party dialogue的研究较多
    two-party dialogue
    • 【AAAI-2017】Discovering Conversational Dependencies between Messages in Dialogs
    • 【AAAI-2019】Learning to Align Question and Answer Utterances in Customer Service Conversation with Recurrent Pointer Networks:https://ojs.aaai.org//index.php/AAAI/article/view/3778
      multi-party dialogue
    • 【NAACL-2018】Learning to Disentangle Interleaved Conversational Threads with a Siamese Hierarchical Network and Similarity Ranking:multi-party, SHCNN, CISIR
    • 【AAAI-2019】A Deep Sequential Model for Discourse Parsing on Multi-Party Dialogues:multi-party
    • 【NAACL-2019】Context-Aware Conversation Thread Detection in Multi-Party Chat:multi-party
    • 【AAAI-2020】Who Did They Respond to? Conversation Structure Modeling Using Masked Hierarchical Transformer:https://ojs.aaai.org/index.php/AAAI/article/view/6524
业界方案二:生成式:(暂不考虑该方案)
  • 论文
    【ACL-2020】Generating Diverse and Consistent QA pairs from Contexts with Information-Maximizing Hierarchical Conditional VAEs:https://www.aclweb.org/anthology/2020.acl-main.20/
    【ACL-2020】Harvesting and Refining Question-Answer Pairs for Unsupervised QA:https://www.aclweb.org/anthology/2020.acl-main.600/
    【EMNLP-2020】QADiscourse - Discourse Relations as QA Pairs: Representation, Crowdsourcing and Baselines:https://www.aclweb.org/anthology/2020.emnlp-main.224/
    【ACL-2020】A Smart System to Generate and Validate Question Answer Pairs for COVID-19 Literature:https://www.aclweb.org/anthology/2020.sdp-1.4.pdf
    【AAAI-2020】On the Generation of Medical Question-Answer Pairs:https://ojs.aaai.org//index.php/AAAI/article/view/6410
    【ACL-2016】QA-It: Classifying Non-Referential It for Question Answer Pairs:https://www.aclweb.org/anthology/P16-3020
    【NAACL-2016】Watson Discovery Advisor: Question-answering in an industrial setting:https://www.aclweb.org/anthology/W16-0101.pdf
    【ACL-2019】Synthetic QA Corpora Generation with Roundtrip Consistency:https://www.aclweb.org/anthology/P19-1620.pdf
    【EMNLP-2017】Question Generation for Question Answering:https://www.aclweb.org/anthology/D17-1090.pdf
    【ACL-2018】Neural Models for Key Phrase Extraction and Question Generation:https://www.aclweb.org/anthology/W18-2609.pdf

############################################################################################

【ECAI-2020】Matching Questions
  • 0
    点赞
  • 1
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值