CoPHE: A Count-Preserving Hierarchical Evaluation Metric in Large-Scale Multi-Label Text Classification
This addresses the need for better evaluation in hierarchical classification tasks like medical coding, but it is incremental as it focuses on metrics rather than model improvements.
The paper tackles the problem of evaluating large-scale multi-label text classification models with hierarchical label spaces, such as ICD-9 coding, by proposing a new depth-based representation and hierarchical evaluation metrics, showing differences in scores compared to standard metrics on MIMIC-III data.
Large-Scale Multi-Label Text Classification (LMTC) includes tasks with hierarchical label spaces, such as automatic assignment of ICD-9 codes to discharge summaries. Performance of models in prior art is evaluated with standard precision, recall, and F1 measures without regard for the rich hierarchical structure. In this work we argue for hierarchical evaluation of the predictions of neural LMTC models. With the example of the ICD-9 ontology we describe a structural issue in the representation of the structured label space in prior art, and propose an alternative representation based on the depth of the ontology. We propose a set of metrics for hierarchical evaluation using the depth-based representation. We compare the evaluation scores from the proposed metrics with previously used metrics on prior art LMTC models for ICD-9 coding in MIMIC-III. We also propose further avenues of research involving the proposed ontological representation.