Exam AWS Certified Machine Learning - Specialty topic 1 question 103 discussion

A. Fastest route must Amazon Services.

A. YES - Rekognition with built-in labels for images & video, Transcribe to convert sound to text and Comprehend for Topic Modeling B. NO - complicated C. NO - complicated D. NO - complicated

AWS services for fastest route

why not C? C. Use Amazon Transcribe to convert speech to text. Use the Amazon SageMaker Neural Topic Model (NTM) and Object Detection algorithms to tag data into distinct categories/classes.

I will choose B，https://aws.amazon.com/cn/getting-started/hands-on/machine-learning-tutorial-label-training-data/

Mechanical Turk is the most accurate, but the three services in letter A is the fastest!

A. Use Amazon Rekognition, Amazon Comprehend, and Amazon Transcribe to tag data into distinct categories/classes is the fastest route to index the assets. These AWS services provide pre-built machine learning models that can be used to tag the content in the archive without the need for building custom models from scratch. This option would be faster than using custom models with the AWS Deep Learning AMI and Amazon EC2 GPU instances, or using Amazon Mechanical Turk for human labeling. Additionally, the use of pre-built models reduces the need for machine learning expertise, aligning with the company's goal of accelerating efforts by its in-house researchers.

A. Option B is for those without ML experience. But "researchers who have limited machine learning expertise“, so A is better.

A. The most straight forward use of services.

I would have said B, but in B it says "label footage" which means it ignored the rest of the data, so i'd go with A

Would go for A

I will go with B

Correct answer is A

B. as no one in-house is an expert and It probably is the fastest way to get there