Authors: ChuanJie Lin ChienWei Pao YenHeng Chen ChiTing Liu HuiHuang Hsu
Publish Date: 2016/08/12
Volume: 40, Issue: 9, Pages: 206-
Abstract
This paper describes the design of an ellipsis and coreference resolution module integrated in a computerized virtual patient dialogue system Real medical diagnosis dialogues have been collected and analyzed Several groups of diagnosisrelated concepts were defined and used to construct rules patterns and features to detect and resolve ellipsis and coreference The best Fscores of ellipsis detection and resolution were 8915 and 8340 respectively The best Fscores of phrasal coreference detection and resolution were 9383 and 8340 respectively The accuracy of pronominal anaphora resolution was 92 for the 3rdperson singular pronouns referring to specific entities and 9731 for other pronounsThis research was funded by the Taiwan Ministry of Science and Technology previous National Science Council grant NSC 1002221E019062 Special thanks go to Dr GeeGwo Yang and Dr ChienHsing Wang in the Hualien Tzu Chi Hospital for their helps in recording diagnosis dialogues and verifying the dataset
Keywords: