SnapToGrid: From Statistical to Interpretable Models for Biomedical Information Extraction

TL;DR

This paper introduces SnapToGrid, a method that converts statistical biomedical information extraction models into interpretable rule-based models, enabling expert editing with minimal performance loss.

Contribution

The paper presents a novel approach to transform statistical models into rule-based models for biomedical extraction, balancing performance and interpretability.

Findings

Small performance penalty when converting models

Experts can easily improve the rule-based model

Interpretable models achieve similar performance to statistical ones

Abstract

We propose an approach for biomedical information extraction that marries the advantages of machine learning models, e.g., learning directly from data, with the benefits of rule-based approaches, e.g., interpretability. Our approach starts by training a feature-based statistical model, then converts this model to a rule-based variant by converting its features to rules, and "snapping to grid" the feature weights to discrete votes. In doing so, our proposal takes advantage of the large body of work in machine learning, but it produces an interpretable model, which can be directly edited by experts. We evaluate our approach on the BioNLP 2009 event extraction task. Our results show that there is a small performance penalty when converting the statistical model to rules, but the gain in interpretability compensates for that: with minimal effort, human experts improve this model to have similar performance to the statistical model that served as starting point.