MLS-C01 Question 143
Single answerA healthcare organization wants to predict whether a patient is at risk of developing diabetes based on various input features, such as age, BMI, family history, and blood test results. The dataset provided is labeled with '0' for non-diabetic patients and '1' for diabetic patients. The organization requires a model with high interpretability to understand the key factors contributing to the predictions. Which model is the most appropriate for this use case?
- A
Linear Regression
- B
Random Forest
- C
Logistic Regression
- D
Convolutional Neural Network (CNN)
Show answer and explanation
Correct answer: C
Explanation
Logistic Regression is the most appropriate choice because it is designed for binary classification problems like predicting diabetes risk and offers high interpretability. This allows the healthcare organization to gain insights into the key factors influencing the predictions, which aligns with their requirement.
- A. Incorrect.
Linear Regression is used for regression tasks where the output is continuous, not suitable for binary classification problems like predicting diabetes.
- B. Incorrect.
Random Forest is a powerful classification model, but it lacks interpretability, which is a key requirement in this use case.
- C. Correct.
Logistic Regression is a suitable choice for binary classification problems and provides high interpretability, allowing the organization to understand the relationship between input features and the prediction.
- D. Incorrect.
Convolutional Neural Networks (CNNs) are designed for image data and complex feature extraction, making them inappropriate for this tabular dataset and the requirement for interpretability.