A company is building a lead-prioritization tool so staff can contact potential customers. Employees must be able to view and manually adjust the importance (weights) of input variables based on their domain expertise. Which type of ML model supports this requirement?

A logistic regression model is ideal because it's a linear model where the weights assigned to each input variable are explicit and easily interpretable. This allows employees to directly view and manually adjust the importance of variables based on their domain expertise, fulfilling the requirement for human intervention. Deep learning models, K-nearest neighbors, and neural networks are generally "black box" models where the internal workings and variable importance are less transparent and not easily modifiable by non-experts. While some techniques exist to interpret these models, direct manual adjustment of individual variable weights is not a standard or practical feature for domain experts.