Optimising Human-Machine Collaboration for Efficient High-Precision Information Extraction from Text Documents

TL;DR

This paper proposes a human-in-the-loop framework for high-precision information extraction from text, combining machine speed with human validation to outperform automated methods in accuracy and efficiency.

Contribution

It introduces a novel framework and tool for weak-supervision labelling with human validation, demonstrating improved precision and efficiency in criminal justice datasets.

Findings

Human-in-the-loop approach achieves manual annotation precision with less time.

Outperforms fully automated baselines in precision.

Effective on criminal justice datasets.

Abstract

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.