On 20th July 2017, I attended IPSJ-SIGNL 232 (情報処理学会 自然言語処理研究発表会) to present our work:
Zeng, Z., Luo, C., Shang L., Li, H., and Sakai, T.: Test Collections and Measures for Evaluating Customer-Helpdesk Dialogues, 情報処理学会研究報告 2017-SIGNL-232, 2017.
It’s my first time to present my research to other researchers, and it made me nervous but was quite exciting.
In our work, we built a test collection which consists of 3700 annotated helpdesk dialogues, and we also proposed a measure to evaluate helpdesk using nuggets than subjective annotation. I received a lot of useful feedback after the presentation.
In additional, I listened to many impressive presentation in this event, e.g., building a chatbot with personality using word embedding.