MLA-C01 Question 144
Single answerA data science team is working on a binary classification problem to predict whether a customer will purchase a product based on their browsing history and demographic information. The dataset is fully labeled and contains 1 million rows with both numerical and categorical features. The team wants to quickly prototype a solution using an Amazon SageMaker built-in algorithm. Which algorithm should they choose?
- A
Linear Learner
- B
K-Means
- C
Factorization Machines
- D
BlazingText
Show answer and explanation
Correct answer: A
Explanation
Linear Learner is the most appropriate SageMaker built-in algorithm for this binary classification problem because it is specifically designed for supervised learning tasks like classification and regression. It efficiently handles large datasets and supports both numerical and categorical features, making it an ideal choice for this scenario.
- A. Correct.
Linear Learner is a good choice for binary classification problems with numerical and categorical features. It supports fast prototyping and scales well with large datasets.
- B. Incorrect.
K-Means is an unsupervised clustering algorithm and is not suitable for labeled binary classification problems.
- C. Incorrect.
Factorization Machines is designed for problems involving sparse datasets, such as recommendation systems, and is less optimal for this fully-labeled binary classification case.
- D. Incorrect.
BlazingText is a word embedding and text classification algorithm and is not applicable to this structured dataset with numerical and categorical features.