Clinical Text Prediction with Numerically Grounded Conditional Language Models

TL;DR

This paper introduces numerically grounded and conditional neural language models for clinical text prediction, significantly enhancing word prediction and completion accuracy by integrating structured knowledge and numerical data.

Contribution

The paper proposes novel grounded and conditional extensions to neural language models that incorporate structured knowledge bases and numerical values, improving clinical text prediction tasks.

Findings

Grounded and conditional models outperform standard models in clinical text prediction.

Combining grounding and conditioning yields the best performance improvements.

Numerical data influence is more prominent at the document level than at individual word prediction.

Abstract

Assisted text input techniques can save time and effort and improve text quality. In this paper, we investigate how grounded and conditional extensions to standard neural language models can bring improvements in the tasks of word prediction and completion. These extensions incorporate a structured knowledge base and numerical values from the text into the context used to predict the next word. Our automated evaluation on a clinical dataset shows extended models significantly outperform standard models. Our best system uses both conditioning and grounding, because of their orthogonal benefits. For word prediction with a list of 5 suggestions, it improves recall from 25.03% to 71.28% and for word completion it improves keystroke savings from 34.35% to 44.81%, where theoretical bound for this dataset is 58.78%. We also perform a qualitative investigation of how models with lower perplexity occasionally fare better at the tasks. We found that at test time numbers have more influence on the document level than on individual word probabilities.