Local Interpretability of Calibrated Prediction Models: A Case of Type 2\n Diabetes Mellitus Screening Test

Résumé fourni par la source

Machine Learning (ML) models are often complex and difficult to interpret due\nto their 'black-box' characteristics. Interpretability of a ML model is usually\ndefined as the degree to which a human can understand the cause of decisions\nreached by a ML model. Interpretability is of extremely high importance in many\nfields of healthcare due to high levels of risk related to decisions based on\nML models. Calibration of the ML model outputs is another issue often\noverlooked in the application of ML models in practice. This paper represents\nan early work in examination of prediction model calibration impact on the\ninterpretability of the results. We present a use case of a patient in diabetes\nscreening prediction scenario and visualize results using three different\ntechniques to demonstrate the differences between calibrated and uncalibrated\nregularized regression model.\n

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.