Skip to content
This repository was archived by the owner on Mar 27, 2022. It is now read-only.
This repository was archived by the owner on Mar 27, 2022. It is now read-only.

Using Customer Service Dialogues for Satisfaction Analysis with Context-Assisted Multiple Instance Learning #137

@tm4roon

Description

@tm4roon

Using Customer Service Dialogues for Satisfaction Analysis with Context-Assisted Multiple Instance Learning

対話ログからカスタマーサービスの満足度予測するタスクを経由し、対話分析が可能な手法を提案している。カスタマーサービスの担当者側の発話における以下のポイントに焦点を当てたモデルを利用し、従来の分類モデルよりも高い分類性能を達成している。

  • 顧客の感情に関係する手がかり(Sentiment Clue): 担当者のどの発話で、顧客の感情が変化したのか?
  • 顧客の発言を補足するための手がかり(Reasoning Clue): 担当者の発話が、顧客の発話のどの部分に対応しているのか?(一度に、たくさん質問された場合などに有効)


また、感情分析に関する教師信号は利用せず、各発話の感情を自動で予測するモデルであるため、対話ログの各発話に感情ラベルを付与せずに、対話の分析が可能なモデルが構築できる。

文献情報

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions