MLS-C01 Question 138
Single answerA financial institution wants to predict potential loan defaults based on customer data, such as income, credit score, and debt-to-income ratio. Which type of machine learning approach should the institution use to address this problem?
- A
Classification
- B
Regression
- C
Clustering
- D
Recommendation
Show answer and explanation
Correct answer: A
Explanation
The problem requires predicting whether a customer will default on a loan, which is a binary classification task (default or no default). Classification models are designed to handle such problems where the output is a discrete category. Other approaches like regression, clustering, or recommendation do not align with the problem requirements.
- A. Correct.
Classification is the correct choice because the problem involves predicting discrete categories (i.e., default or no default). Classification models are well-suited for binary or multi-class problems where the outcome belongs to distinct classes.
- B. Incorrect.
Regression is incorrect because regression models predict continuous numeric values, such as sales or temperature, rather than discrete categories like loan default status.
- C. Incorrect.
Clustering is incorrect because clustering is an unsupervised learning technique used to group similar instances without labeled outcomes, not for predicting predefined categories.
- D. Incorrect.
Recommendation is incorrect because recommendation systems are used to suggest items or content to users based on preferences or past behavior, not to predict categories like loan default.