DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset

TLDR

DailyDialog contains human‑written, low‑noise dialogues that reflect everyday communication across diverse daily life topics. The authors aim to create a high‑quality multi‑turn dialogue dataset and evaluate existing dialog systems on it to advance the field. The dataset is manually annotated with communication intention and emotion labels. DailyDialog is publicly available at http://yanran.li/dailydialog.

Abstract

We develop a high-quality multi-turn dialog dataset, DailyDialog, which is intriguing in several aspects. The language is human-written and less noisy. The dialogues in the dataset reflect our daily communication way and cover various topics about our daily life. We also manually label the developed dataset with communication intention and emotion information. Then, we evaluate existing approaches on DailyDialog dataset and hope it benefit the research field of dialog systems. The dataset is available on http://yanran.li/dailydialog