A company needs an interpretable machine learning model to evaluate loan application risk. Which model or algorithm choice best satisfies the requirement for interpretability?

Logistic regression is a highly interpretable model because its coefficients directly show the impact and direction of each input feature on the predicted probability. For loan applications, this means you can clearly see how factors like credit score or income influence the risk assessment, which is crucial for regulatory compliance and explaining decisions to applicants. Deep learning models are generally considered "black boxes" due to their complex, non-linear structures, making it difficult to understand individual feature contributions. K-means clustering is an unsupervised learning algorithm used for grouping data, not for predictive modeling or risk assessment. Random Cut Forest is an unsupervised anomaly detection algorithm, not suitable for interpretable risk prediction.