PRODUCTION SCENARIO
A support team wants to group incoming tickets by similar meaning before anyone has labeled the categories.
Which approach will meet these requirements?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Clustering embeddings of the incoming ticket text
Embeddings place tickets that mean similar things near each other, and clustering then finds those groups without anyone defining the categories first. Amazon Comprehend custom classification organizes documents into your own categories, but you supply labeled examples of each category, which is the step that has not happened. Key phrase extraction describes what is in one ticket and never compares two tickets. Summarizing shortens each ticket and still leaves them ungrouped.
Official Source: https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html ↗