Exam UiPath-SAIv1 Topic 1 Question 21 Discussion
Actual exam question for UiPath's UiPath-SAIv1 exam
Question #: 21
Topic #: 1
Question #: 21
Topic #: 1
While training a UiPath Communications Mining model, the Search feature was used to pin a certain label on a few communications. After retraining, the new model version starts to predict the tagged label but infrequently and with low confidence.
According to best practices, what would be the correct next step to improve the model's predictions for the label, in the "Explore" phase of training?
According to best practices, what would be the correct next step to improve the model's predictions for the label, in the "Explore" phase of training?
Suggested Answer: B Vote an answer
According to the UiPath documentation, the 'Teach' training mode is used to improve the model's predictions for a specific label by pinning it to more communications that match the label's criteria. This helps the model learn from more examples and increase its confidence and accuracy. The 'Teach' mode also allows you to unpin the label from communications that do not match it, which helps the model avoid false positives. The other training modes are not as effective for this purpose, as they either focus on different aspects of the model performance or do not provide enough feedback to the model.
References:
* Model training and labelling best practice
* Overview of the model training process
* Model Training FAQs
References:
* Model training and labelling best practice
* Overview of the model training process
* Model Training FAQs
by Molly at Apr 23, 2025, 07:33 AM
Contact Us
If you have any question please leave me your email address, we will reply and send email to you in 12 hours.
Our Working Time: ( GMT 0:00-15:00 ) From Monday to Saturday
Support: Contact now
Comments
Upvoting a comment with a selected answer will also increase the vote count towards that answer by one. So if you see a comment that you already agree with, you can upvote it instead of posting a new comment.
Report Comment
Commenting
You can sign-up / login (it's free).