AI-102 Question 277
Select 3You are building a conversational AI chatbot using Azure Language Understanding (LUIS) to handle customer support queries. During testing, you notice that the model often misinterprets user intents, especially when users use synonyms or slightly different phrasing. What steps should you take to optimize the language understanding model?
- A
Add more example utterances for the intents in your training dataset.
- B
Enable active learning and review suggestions for new utterances.
- C
Combine multiple intents into one to simplify the model's structure.
- D
Use Azure Monitor to track the model’s response times and adjust latency settings.
- E
Add synonyms and phrase variations to the entity list in the LUIS model.
Show answer and explanation
Correct answers: A, B, E
Explanation
To optimize a language understanding model, it is crucial to improve the quality and diversity of the training data by adding more example utterances and leveraging active learning to capture real-world user inputs. Additionally, defining synonyms and phrase variations for entities ensures the model can handle different ways users express the same concept. These steps collectively enhance the model’s ability to correctly interpret user intents and entities.
- A. Correct.
Adding more example utterances helps the model better understand the ways users might express the same intent, improving accuracy and reducing misinterpretations.
- B. Correct.
Enabling active learning allows you to review and incorporate user queries that the model struggles with, helping refine intent recognition over time.
- C. Incorrect.
Combining multiple intents into one is not a valid optimization strategy, as it can reduce clarity and make the model less accurate in distinguishing between different user intents.
- D. Incorrect.
Adjusting latency settings in Azure Monitor does not address the issue of misinterpreted user intents, as it is focused on performance rather than accuracy.
- E. Correct.
Adding synonyms and phrase variations to the entity list helps the model recognize different ways users might refer to the same concept, improving its language understanding.