Which solution should the ML engineer apply to perform multi-class text classification of the user feedback?
Use the SageMaker Latent Dirichlet Allocation (LDA) algorithm.
Use the SageMaker BlazingText algorithm.
Use the SageMaker Neural Topic Model (NTM) algorithm.
Use the SageMaker CatBoost algorithm.
Explanations:
The SageMaker Latent Dirichlet Allocation (LDA) algorithm is a topic modeling algorithm, not suitable for supervised classification tasks like the one described. It is primarily used to identify topics in a collection of text but does not perform the multi-class classification required for the user feedback categorization.
The SageMaker BlazingText algorithm is a highly optimized algorithm for text classification tasks, especially suited for large datasets. It supports multi-class classification and works well for problems like user feedback categorization, where the goal is to classify text into predefined categories.
The SageMaker Neural Topic Model (NTM) algorithm is designed for topic modeling and unsupervised learning tasks, similar to LDA. It helps discover topics within a corpus of text but is not intended for supervised classification tasks with fixed categories.
The SageMaker CatBoost algorithm is a gradient boosting algorithm that excels with structured data. It is not well-suited for text classification tasks, as it requires the text data to be preprocessed into a structured format, making it less effective than a dedicated text classification algorithm like BlazingText.