SOL-C01 · Question #203
SOL-C01 Question #203: Real Exam Question with Answer & Explanation
The correct answer is B: Preprocess the customer feedback using a sentiment analysis UDF that identifies and flags. Option B is the most practical and effective strategy. While CLASSIFY_TEXT is a black-box function, you can improve its results by providing better input. Using sentiment analysis to flag sarcastic comments allows you to 'hint' at the true sentiment and improves overall accuracy.
Question
A data analyst is using Snowflake Cortex's CLASSIFY TEXT function to categorize customer feedback. They notice that some feedback containing sarcasm is consistently misclassified. Which of the following strategies would be MOST effective in improving the accuracy of CLASSIFY TEXT in this scenario, considering the function's limitations and capabilities?
Options
- AManually create a custom classification model within Snowflake using Python User-Defined
- BPreprocess the customer feedback using a sentiment analysis UDF that identifies and flags
- CThere is no way to improve the accuracy of CLASSIFY _ TEXT. Its performance is fixed and
- DSince CLASSIFY TEXT is a black-box function, the only way to improve performance is to provide
- ESubmit a feature request to Snowflake support requesting they improve the CLASSIFY TEXT
Explanation
Option B is the most practical and effective strategy. While CLASSIFY_TEXT is a black-box function, you can improve its results by providing better input. Using sentiment analysis to flag sarcastic comments allows you to 'hint' at the true sentiment and improves overall accuracy. Option A involves a custom model, which is outside the scope of using CLASSIFY _ TEXT. Options C, D, and E are incorrect due to understating the possible improvement strategies or relying on uncontrollable external factors.
Topics
Community Discussion
No community discussion yet for this question.