Exam AWS Certified Machine Learning - Specialty topic 1 question 154 discussion

C and E B needs to build custom model

It asked 2 answers, but I can see only one answer. Please advise. Thanks!

(Choose two.) C,E

C E - Amazon Translate can handle the translation from Spanish to English, ensuring all comments are in a single language. Amazon Comprehend provides robust topic modeling capabilities to identify the most discussed topics in the translated comments. Use Amazon Translate to translate from Spanish to English, if necessary. Use Amazon SageMaker Neural Topic Model (NTM) to find the topics. Amazon SageMaker Neural Topic Model (NTM) is an unsupervised learning algorithm designed for topic modeling, which can effectively identify topics in the translated comments

CE are the answers. C use Amazon Comprehend for topic modeling - use LDA (https://docs.aws.amazon.com/comprehend/latest/dg/topic-modeling.html) E is using NTM (https://docs.aws.amazon.com/sagemaker/latest/dg/ntm.html) For ease of use, I will start with Amazon Comprehend and go to NTM if the success criteria isn't met.

C, E is the correct answer.

C & E based on comments, but you are not allowed to select multiple choices.

C and E - Use translate so that text is in common language - In options with translate only Comprehend and NTM allow for topic modeling (C & E) Other options Blazingtext is for text classification, not topic modelling, LDA is requireres user specified topics and Lex is for conversational interfaces

C & E Option C (Amazon Translate and Amazon Comprehend): This is a strong combination. Amazon Translate can be used to translate Spanish comments into English, and then Amazon Comprehend, which supports topic modeling, can be used to identify the most discussed topics. Option E (Amazon Translate and Amazon SageMaker Neural Topic Model): This is also a viable combination. Amazon Translate would handle the translation of Spanish comments, and the Neural Topic Model (NTM) in Amazon SageMaker can then be used for topic modeling. NTM uses neural networks for topic discovery and is well-suited for analyzing large sets of text data.

B and E I dont think amazon comprehend can do topic modelling. LDA is used for topic modelling

BCE are all right. https://docs.amazonaws.cn/en_us/sagemaker/latest/dg/algos.html LDA and NTM are all topic modeling tools.

A. NO - BlazingText is word2vec, will not do topic modeling alone B. NO - Translate better than custom seq2seq C. NO - NTM better than LDA used by Comprehend D. NO - Lex is for chatbots E. YES

The right answers are C & E The other steps are not suitable because: A. The BlazingText algorithm is for word embeddings and text classification, not topic modeling. B. The LDA algorithm is an unsupervised learning algorithm that requires a user-specified number of topics. D. Amazon Lex is for building conversational interfaces, not extracting topics from content

Correct answer is BE

It has to be B + C .. for spanish to English use Translate. For Topics it has to be LDA.

For me: B - C - E are correct: it's solved translation + topic modelling. The question is not well construct from my POV.

C and E