Optimising Human-Machine Collaboration for Efficient High-Precision Information Extraction from Text Documents
This addresses the need for efficient and reliable information extraction in high-stakes applications like criminal justice, though it is incremental as it builds on existing human-machine collaboration concepts.
The paper tackles the problem of achieving high-precision information extraction from text documents by proposing a human-in-the-loop framework that combines weak-supervision labeling with human validation, resulting in precision comparable to manual annotation with reduced time and outperforming fully automated baselines.
While humans can extract information from unstructured text with high precision and recall, this is often too time-consuming to be practical. Automated approaches, on the other hand, produce nearly-immediate results, but may not be reliable enough for high-stakes applications where precision is essential. In this work, we consider the benefits and drawbacks of various human-only, human-machine, and machine-only information extraction approaches. We argue for the utility of a human-in-the-loop approach in applications where high precision is required, but purely manual extraction is infeasible. We present a framework and an accompanying tool for information extraction using weak-supervision labelling with human validation. We demonstrate our approach on three criminal justice datasets. We find that the combination of computer speed and human understanding yields precision comparable to manual annotation while requiring only a fraction of time, and significantly outperforms fully automated baselines in terms of precision.