A business analytics professional wants to predict the likelihood of patients to have a second heart attack using their past data. Which of the following models should s/he use?

• A. Linear Regression
• B. K-Means
• C. Logistic Regression
• D. Association Rule

Answer: (C)

When the outcome is binary (whether a patient has a second heart attack), you want a model that outputs a probability for that event and handles a yes/no target. Logistic regression does exactly that by modeling the log-odds of the event as a linear function of the predictors and then applying the logistic function to produce a probability between 0 and 1. This makes the results interpretable in terms of risk and allows you to include various patient features (continuous or categorical) while keeping predicted probabilities valid.

Linear regression isn’t ideal here because it can predict values outside the 0–1 range and relies on assumptions that don’t hold for a binary outcome. K-means is an unsupervised clustering method that groups similar observations rather than predicting the probability of an event for new individuals. Association rule mining seeks to discover patterns and relationships, not to provide individual-level probability predictions for a medical outcome.