Dialogue-Based Relation Extraction
This work addresses the problem of extracting relations from multi-turn dialogues, which is incremental as it adapts existing RE methods to a new conversational context.
The authors introduced DialogRE, the first human-annotated dataset for relation extraction in dialogues, and showed that adding speaker-aware information to the best-performing model improved performance in both standard and conversational evaluation settings.
We present the first human-annotated dialogue-based relation extraction (RE) dataset DialogRE, aiming to support the prediction of relation(s) between two arguments that appear in a dialogue. We further offer DialogRE as a platform for studying cross-sentence RE as most facts span multiple sentences. We argue that speaker-related information plays a critical role in the proposed task, based on an analysis of similarities and differences between dialogue-based and traditional RE tasks. Considering the timeliness of communication in a dialogue, we design a new metric to evaluate the performance of RE methods in a conversational setting and investigate the performance of several representative RE methods on DialogRE. Experimental results demonstrate that a speaker-aware extension on the best-performing model leads to gains in both the standard and conversational evaluation settings. DialogRE is available at https://dataset.org/dialogre/.