Exam AWS Certified Big Data - Specialty topic 1 question 20 discussion

Answer is A. Key is "Redshift has the past two years of historical data". This means we have labelled data that we can use to train a linear regression model to predict RCU and WCU.

my selection A

"needs to calculate how much read and write throughput should be provisioned for DynamoDB table for each week in advance" A regression model is more suitable for this job, you can just run 2 models, one for WCU, one for RCU. B is for detecting anomalies.

Normally, when predict some numbers like salary, price we use the regression model based on the histrory data. So personally, i agree with the B as it is used to predicted the number of the read and write throughput

The correct answer is B ... An anomaly score with low values indicates that the data point is considered “normal” whereas high values indicate the presence of an anomaly. The definitions of “low” and “high” depend on the application, but common practice suggests that scores beyond three standard deviations from the mean score are considered anomalous. https://aws.amazon.com/blogs/machine-learning/use-the-built-in-amazon-sagemaker-random-cut-forest-algorithm-for-anomaly-detection/

Not sure about A, mainly because regression is to precit just one numeric value, not two (RCU/WCU). Regarding Random forest (B) is also used for reggresion tasks as suggest here: https://www.newgenapps.com/blog/random-forest-analysis-in-ml-and-when-to-use-it

A. Feed the data into Amazon Machine Learning and build a regression model is the right answer because regression model is used for numeric value and here we are looking for RCU/WCU which is numeric as well. though B can be an option but it's not cost effective as it needs EMR cluster

it should be A .. Random forest in B is for classification model.

A I think

A, because historiccal using regression ml model

Why not B?

A is correct one

Correct. A

a? anyone?