Exam AWS Certified Machine Learning - Specialty topic 1 question 149 discussion

D is the correct answer. Following from the previous comment. The company wants to minimize the infrastructure and data science resources needed to evaluate the messages. Therefore any custom services would be eliminated (A and B). Similarly DynamoDB would add complexity to the infrastructure there C is eliminated. leaving D

Recording Sentiment in cloudwatch metric seems odd. DynamoDB seems more accurate.

Option D is the right answer. Following are the key terms in question to notice, sentiment expressed in social media posts --> Comprehend configure alarms based on various thresholds --> CloudWatch (can send alerts without SNS) wants to minimize the infrastructure and data science resources --> AWS S3

The best services for the data science team to use to deliver this solution are option D, trigger an AWS Lambda function when social media posts are added to the S3 bucket, call Amazon Comprehend for each post to capture the sentiment in the message and record the sentiment in a custom Amazon CloudWatch metric and in S3, and use CloudWatch alarms to notify analysts of trends. By doing so, the data science team can use Amazon Comprehend, a natural language processing (NLP) service that uses machine learning to find insights and relationships in text, to evaluate the sentiment expressed in social media posts. Amazon Comprehend can detect positive, negative, neutral, or mixed sentiment from text input. The data science team can also use AWS Lambda, a service that lets you run code without provisioning or managing servers, to trigger a function when posts are added to the S3 bucket and call Amazon Comprehend for each post.

Amazingly D is possible - https://catalog.us-east-1.prod.workshops.aws/workshops/4faab440-8c3a-4527-bd11-0c88a6e6213c/en-US/30-build-the-application/400-send-sentiment-to-cloudwatch I was so sure of option C, because sending a sentiment to a custom CloudWatch metric just didn't make any sense. But you learn something new everyday.

This is a puzzling question, as both answers C and D miss essential steps: C is missing DynamoDB Streams to capture new records D is missing a notification mechanism like SNS, as CloudWatch Alarms alone can only be used as a trigger, but are not sufficient for notification I agree that A and B should be eliminated for requiring data science development

I also do agree that D is correct answer. In A, why we are adding extra dependency of Dynamo DB.

D, blazing text is not for sentiment analysis. The Amazon SageMaker BlazingText algorithm provides highly optimized implementations of the Word2vec and text classification algorithms. The Word2vec algorithm is useful for many downstream natural language processing (NLP) tasks, such as sentiment analysis, named entity recognition, machine translation, etc. Text classification is an important task for applications that perform web searches, information retrieval, ranking, and document classification.