Real-time On-Demand Crowd-powered Entity Extraction
This work addresses the need for fast and reliable entity extraction in dialogue systems, though it is incremental as it builds on existing output-agreement mechanisms.
The paper tackled the problem of extracting key information from user utterances in interactive systems by proposing a time-limited output-agreement mechanism, achieving high-quality results with an average response time under 9 seconds in experiments on the ATIS dataset.
Output-agreement mechanisms such as ESP Game have been widely used in human computation to obtain reliable human-generated labels. In this paper, we argue that a "time-limited" output-agreement mechanism can be used to create a fast and robust crowd-powered component in interactive systems, particularly dialogue systems, to extract key information from user utterances on the fly. Our experiments on Amazon Mechanical Turk using the Airline Travel Information System (ATIS) dataset showed that the proposed approach achieves high-quality results with an average response time shorter than 9 seconds.