In data mining, association rules are commonly used to generate which type of outputs?

• A. Predictions
• B. Models
• C. Recommendations
• D. Insights

Answer: (C)

Association rules reveal relationships between items and are used to decide what to show or suggest next. The outputs they produce are recommendations, because the rules indicate which items are commonly bought together, allowing a system to suggest additional items when a customer buys or views something. For example, if the rule shows that people who buy bread often buy butter, the system can recommend butter when bread is in the cart. The strength of these suggestions is quantified by metrics like support, confidence, and lift, which help determine which recommendations to prioritize. While these rules can provide insights about item co-occurrence, they are most directly used to generate practical recommendations rather than single-item predictions or a traditional predictive model.