Exam AWS Certified Machine Learning - Specialty topic 1 question 108 discussion

I think the answer is D - RCF works together with Data Analytics, and sliding window helped on new information

D uses the built-in RCF algorithm, which is designed for anomaly detection on streaming data and can adapt to changing patterns over time.It does not require any training data or preprocessing steps, as the RCF algorithm can learn from the streaming data directly. It uses a sliding window, which allows for continuous updating of the anomaly scores based on the most recent data points. It leverages the Amazon Kinesis Data Analytics service, which provides a scalable and managed platform for running SQL queries on streaming data. Option A requires training an RCF model on historic data, which may not reflect the current web traffic patterns. It also adds complexity and latency by invoking a Lambda function for each record.

Answer D

Letra B está descartada, pois trás um modelo supervisionado de classificação para um problema não supervisionado. Letra C trás outro modelo que não é recomendado também, em comparação ao RCF. A solução mais fácil de implementar e que atinge os critérios pedidos é a Letra D. Letra A está errada, pois usamos KDS para ingestão apenas.

the data scientist needs to identify unusual web traffic patterns in real-time and adapt to changing web patterns over time. Amazon Kinesis Data Analytics provides real-time analytics capabilities on streaming data. The Amazon Random Cut Forest (RCF) SQL extension is designed for anomaly detection in streaming data, which fits the requirement to calculate an anomaly score for each web traffic entry.

Answer is D "The algorithm starts developing the machine learning model using current records in the stream when you start the application. The algorithm does not use older records in the stream for machine learning, nor does it use statistics from previous executions of the application." https://docs.aws.amazon.com/kinesisanalytics/latest/sqlref/sqlrf-random-cut-forest.html

D it is. easy one

RCF is dynamic and adapts with time. D seems more appropriate.

It is A. The only way to handle the historic data is using sagemaker and you can preprocess a data stream using a lambda.

Definitly D, Data Anaytics is using RCF, Using window for selecting data with SQL

One more reason to select D, not A, is there is no Lambda function to preprocess record in Kinesis Data Stream.

“ Adapt unusual event identification to changing web patterns over time.” -> option A does not satisfy this, only mentions build the model once

The data scientist has access to unlabeled historic data to use, if needed. D has no mention of this. Also, A says the lambda function provides data enrichment. For me it's A.

A and D both seems to works. But A does not satisfy requirement 2, adapt to patterns over time. Since the model is only trained on old data. So D may be better.

It is definitely D