AI-900 Question 193
Select 2A company wants to improve its customer support by analyzing customer reviews to identify the main topics of complaints and detect the sentiment behind them. Which NLP workload scenarios are most suitable for this task?
- A
Named Entity Recognition (NER)
- B
Sentiment Analysis
- C
Topic Modeling
- D
Speech-to-Text Conversion
- E
Machine Translation
Show answer and explanation
Correct answers: B, C
Explanation
To analyze customer reviews effectively, the company needs to understand both the sentiment expressed and the main topics of concern. Sentiment Analysis helps identify the emotional tone behind a review, while Topic Modeling identifies the recurring themes or topics. These two NLP workload scenarios are most relevant to the given task.
- A. Incorrect.
Named Entity Recognition (NER) focuses on identifying specific entities like names, dates, or locations in text. While useful in some scenarios, it is not directly relevant to identifying complaint topics or customer sentiment.
- B. Correct.
Sentiment Analysis is used to determine the emotional tone or sentiment (e.g., positive, negative, neutral) expressed in text. This is essential for analyzing customer reviews to detect complaints.
- C. Correct.
Topic Modeling is suitable for identifying recurring themes or topics in large amounts of text, making it highly relevant for discovering the main topics of customer complaints.
- D. Incorrect.
Speech-to-Text Conversion is used to convert spoken language into written text. While it is useful in other scenarios, it is not relevant for analyzing text-based customer reviews.
- E. Incorrect.
Machine Translation is used to translate text between different languages. It is not directly related to detecting sentiment or identifying complaint topics in customer reviews.