Exam AWS Certified Machine Learning - Specialty topic 1 question 260 discussion

D makes more sence to me. Collaborative filtering takes into account other users preferences which is what we want to avoid because we do not want irrelevant promotions

Answer 'C' is better for efficiency

As per the documentation: https://docs.aws.amazon.com/sagemaker/latest/dg/fact-machines.html Factorization machines are a good choice for tasks dealing with high dimensional sparse datasets, such as click prediction and item recommendation. And factorization machines is better with sparse data.

C. Use the Neural Collaborative Filtering algorithm with a SageMaker batch inference job This solution uses the Neural Collaborative Filtering algorithm to leverage the latest techniques in recommendation systems, while SageMaker's batch inference jobs provide efficient and cost-effective processing of recommendations in bulk. This aligns well with the company's weekly email campaigns and minimizes operational overhead.

I will go with D, as it is more operationally eficient

A. NO - Factorization Machines is classification B. NO - an endpoint needed be invoked C. YES - Collaborative Filtering is good for recommendations based on past activities, and a batch job will generate the fiel we want D. NO - Factorization Machines is classification

My choice D

factorization machine algorithm is used for regression or classification. Generating recommendation is neither. Use neural collaborative filtering and do batch inference to identify email addresses.

From Chat GPT The solution that will meet the requirements with the MOST operational efficiency is option C: Use the Neural Collaborative Filtering algorithm to build a model that can generate personalized offer recommendations for customers. Deploy a SageMaker batch inference job to generate offer recommendations. Feed the offer recommendations into the bulk email marketing system. By using the Neural Collaborative Filtering algorithm, the ML team can build a model that can provide personalized offer recommendations based on customer profiles and past accepted offers. Deploying a SageMaker batch inference job allows for efficient processing of a large batch of customer data to generate offer recommendations. These recommendations can then be fed directly into the bulk email marketing system, streamlining the process and improving operational efficiency.

: Use the Factorization Machines algorithm to build a model that can generate personalized offer recommendations for customers. Deploy a SageMaker batch inference job to generate offer recommendations. Feed the offer recommendations into the bulk email marketing system

C batch predictions and collaborative

C is a better option for efficiency.