Three Options to Improve Prediction Accuracy

Correct Answers: A, B and D.

Option A is correct.

You can review dashboard analysis to find unclear predictions marked with Orange color and incorrect predictions marked with Red color.

Option B is correct.

You can log user queries by enabling active learning, reviewing the utterances with lower prediction scores and review/modify the intents and entities to improve the model accuracy.

Option C is incorrect because the value of the endpoint query parameter log has to be maintained as true to log user queries.

Option D is correct.

In the LUIS portal, you can use a pattern to improve prediction accuracy for utterances.

Option E is incorrect because adding more example utterances would not fix the utterances with incorrect or unclear predictions.

References:

To learn more about improving the prediction accuracy of a LUIS application, use the links given below:

In the given scenario, the top intent and next intent scores are close enough, and there are a few utterances that are not predicted for the labeled intent. To improve the prediction accuracy, there are several options available. Three possible options are:

A. Review dashboard colors and find intents with incorrect/unclear predictions: The LUIS dashboard provides a visual representation of the performance of your LUIS application. By reviewing the dashboard colors, you can quickly identify which intents are performing well and which are not. You can then focus on the poorly performing intents and review the example utterances to improve the prediction accuracy.

B. Enable active learning, capture endpoint queries and relabel entities: Active learning is a feature in LUIS that allows you to capture endpoint queries and relabel entities. By enabling active learning, LUIS will learn from the queries it receives and suggest new example utterances to improve the prediction accuracy. You can then review and relabel these new example utterances to further improve the accuracy.

C. Set endpoint query parameter log=false, capture endpoint queries and relabel entities: Setting the endpoint query parameter log=false will disable logging of the endpoint queries. By capturing the endpoint queries and relabeling entities, you can further improve the prediction accuracy by providing LUIS with more accurate and relevant data.

D. Add example utterances as a pattern, train and publish the application again: Adding example utterances as a pattern can help improve the prediction accuracy by teaching LUIS to recognize specific patterns in the user's input. After adding the new examples, you need to train and publish the application again to apply the changes.

E. Add more example utterances to improve prediction score: Adding more example utterances can help LUIS understand the user's intent better and improve the prediction accuracy. By providing LUIS with a larger and more diverse set of example utterances, it can learn to recognize a wider range of user input and provide more accurate predictions.

In conclusion, to improve the prediction accuracy of a LUIS application, you can review the dashboard colors, enable active learning, capture endpoint queries, relabel entities, add example utterances as a pattern, and add more example utterances. The best approach depends on the specific scenario and the type of errors being made by LUIS. It's recommended to try multiple approaches and continually monitor the performance to achieve the best results.