AWS Certified Machine Learning – Specialty — Question 333
An ecommerce company has observed that customers who use the company's website rarely view items that the website recommends to customers. The company wants to recommend items to customers that customers are more likely to want to purchase.
Which solution will meet this requirement in the SHORTEST amount of time?
Answer options
- A. Host the company's website on Amazon EC2 Accelerated Computing instances to increase the website response speed.
- B. Host the company's website on Amazon EC2 GPU-based instances to increase the speed of the website's search tool.
- C. Integrate Amazon Personalize into the company's website to provide customers with personalized recommendations.
- D. Use Amazon SageMaker to train a Neural Collaborative Filtering (NCF) model to make product recommendations.
Correct answer: C
Explanation
Amazon Personalize is a fully managed service that allows developers to quickly build and deploy real-time, customized recommendations with minimal machine learning experience, making it the fastest solution. In contrast, using Amazon SageMaker to build and train a custom Neural Collaborative Filtering (NCF) model requires significant time and effort for data preparation, model training, and deployment. Upgrading EC2 instance types (Options A and B) only addresses infrastructure performance and does not improve the quality or relevance of the recommendations.