MLS-C01 · Question #203
A retail company wants to update its customer support system. The company wants to implement automatic routing of customer claims to different queues to prioritize the claims by category. Currently…
The correct answer is D. Export the database to a .csv file with two columns: claim_label and claim_text. Use. The correct solution exports the database to a CSV with two columns - claim_label (the category an operator assigned) and claim_text (the claim content) - and uses Amazon Comprehend Custom Classification. Comprehend Custom Classification is a no-code/low-code managed service…
Question
A retail company wants to update its customer support system. The company wants to implement automatic routing of customer claims to different queues to prioritize the claims by category. Currently, an operator manually performs the category assignment and routing. After the operator classifies and routes the claim, the company stores the claim’s record in a central database. The claim’s record includes the claim’s category. The company has no data science team or experience in the field of machine learning (ML). The company’s small development team needs a solution that requires no ML expertise. Which solution meets these requirements?
Options
- AExport the database to a .csv file with two columns: claim_label and claim_text. Use the
- BExport the database to a .csv file with one column: claim_text. Use the Amazon
- CUse Amazon Textract to process the database and automatically detect two columns:
- DExport the database to a .csv file with two columns: claim_label and claim_text. Use
How the community answered
(66 responses)- A14% (9)
- B9% (6)
- C3% (2)
- D74% (49)
Explanation
The correct solution exports the database to a CSV with two columns - claim_label (the category an operator assigned) and claim_text (the claim content) - and uses Amazon Comprehend Custom Classification. Comprehend Custom Classification is a no-code/low-code managed service that trains a custom text classifier using your labeled data. It requires no ML expertise: you upload the CSV, trigger training, and then call the API endpoint for inference. This perfectly matches the requirements: the company has labeled historical data (claim_label exists in the database), no ML expertise, and a small dev team. Option B with only one column (claim_text, no labels) would use unsupervised topic modeling, which cannot learn the specific categories the operators use. Option C using Amazon Textract is designed for document OCR and data extraction, not for training classifiers. Option A may be similar but differ in a critical configuration detail. The labeled data + Comprehend Custom Classification combination is the minimal-expertise path to automated categorization.
Topics
Community Discussion
No community discussion yet for this question.