AI-900: Enhancing User Experience: Improving a QnA Maker Bot with Professional Interactions

Discover effective strategies to enhance user-friendliness in a QnA Maker bot built with a frequently asked questions (FAQ) page. Learn how to add professional greetings and responses that optimize user engagement. Explore techniques such as increasing confidence thresholds, enabling active learning, creating multi-turn questions, and incorporating chit-chat functionality. Elevate your QnA Maker bot to deliver seamless and informative interactions to users.

Question

You build a QnA Maker bot by using a frequently asked questions (FAQ) page. You need to add professional greetings and other responses to make the bot more user friendly. What should you do?

A. Increase the confidence threshold of responses.
B. Enable active learning.
C. Create multi-turn questions.
D. Add chit-chat.

Answer

D. Add chit-chat.

Explanation

The correct answer is D. Add chit-chat. Chit-chat is the feature that allows you to add a pre-populated set of conversational responses to your QnA Maker bot, based on different personalities such as professional, friendly, witty, caring, or enthusiastic. Chit-chat can help to make your bot more user friendly, engaging, and natural, by adding greetings, small talk, humor, and other expressions to your bot’s responses. You can add chit-chat to your QnA Maker bot during the knowledge base creation, or to an existing knowledge base, by choosing one of the personality datasets provided by QnA Maker. You can also edit or customize the chit-chat questions and answers to suit your bot’s voice and tone.

The other options are incorrect because:

• Increasing the confidence threshold of responses is not a way to make the bot more user friendly, but rather a way to control the quality and accuracy of the bot’s responses. The confidence threshold is the minimum score that a response must have to be returned by the bot. If no response meets the threshold, the bot will return a default message such as “No good match found in KB”. Increasing the confidence threshold can help to reduce the number of irrelevant or incorrect responses, but it can also increase the number of unanswered queries.
• Enabling active learning is not a way to make the bot more user friendly, but rather a way to improve the bot’s performance and intelligence over time. Active learning is the feature that allows you to review and label the user queries that the bot is not confident about, and add them to the knowledge base as new questions and answers. Active learning can help to expand the bot’s coverage and relevance, and reduce the knowledge gaps in the bot’s responses.
• Creating multi-turn questions is not a way to make the bot more user friendly, but rather a way to handle complex or follow-up queries from the user. Multi-turn questions are the questions that require more than one interaction between the user and the bot to be answered. For example, if the user asks “How do I reset my password?”, the bot might ask a follow-up question such as “What is your email address?”. Multi-turn questions can help to make the bot more conversational and contextual, and provide more precise and personalized responses.

References

Microsoft Docs > Azure > Cognitive Services  >QnA Maker > Azure Cognitive Services QnA Maker

Microsoft Azure AI Fundamentals AI-900 certification exam practice question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the Microsoft Azure AI Fundamentals AI-900 exam and earn Microsoft Azure AI Fundamentals AI-900 certification.