Attitude detection for one-round conversation: Jointly extracting target-polarity pairs

Zhaohao Zeng, Ruihua Song, Pingping Lin, Tetsuya Sakai

研究成果: Article査読

抄録

We tackle Attitude Detection, which we define as the task of extracting the replier’s attitude, i.e., a target-polarity pair, from a given one-round conversation. While previous studies considered Target Extraction and Polarity Classification separately,we regard them as subtasks of Attitude Detection. Our experimental results show that treating the two subtasks independently is not the optimal solution for Attitude Detection, as achieving high performance in each subtask is not sufficient for obtaining correct target-polarity pairs. Our jointly trained model AD-NET substantially outperforms the separately trained models by alleviating the target-polarity mismatch problem. By employing pointer networks to consider the target extraction task a boundary prediction problem instead of a sequence labelling problem, the model obtained better performance and faster training/inference than LSTM and LSTM-CRF based models. Moreover, we proposed a method utilising the attitude detection model to improve retrieval-based chatbots by re-ranking the response candidates with attitude features. Human evaluation indicates that with attitude detection integrated, the new responses to the sampled queries are statistically significantly more consistent, coherent, engaging and informative than the original ones obtained from a commercial chatbot.

本文言語English
ページ(範囲)742-751
ページ数10
ジャーナルJournal of information processing
27
DOI
出版ステータスPublished - 2019

ASJC Scopus subject areas

  • Computer Science(all)

フィンガープリント 「Attitude detection for one-round conversation: Jointly extracting target-polarity pairs」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル