by Roboyo
1
ML Model
135
Summary
Summary
This AI Center model will classify emails according to their contents based on buckets that are given to it by supplying appropriate training data.
Overview
Overview
The downloadable skill and ML model which was developed by Roboyo in cooperation with DatasparQ will provide a good base level of performance for emails/text in one language (English or German).
The model needs to be trained before first use and the “buckets” the model must use to classify the given emails/texts need to be clear and supplied in the training set. After the deployment the skill can be used directly in UiPath Studio to classify any text supplied as a string - no fiddling around with JSON or REST calls needed!
The model gives back a confidence score based on which you can decide on how to proceed with the classification. Is the confidence too low? Send it to the action center and let a human classify the correct bucket and feed the results back for retraining the model!
This model is a great starting point for using AI Center to sort and classify text directly out of the Studio and then through your robots with ease! However, we understand that real-world use cases are often more complicated. Emails come in different languages and may require mapping to hierarchical classifications. Additionally, the measure of success for a classifier will vary across different use cases.
If you have further needs for this model, please get in touch with us by opening a question on this listing page and we will get back to you!
Features
Features
Ready-to-use ML model specifically built on AI Center to make the start into the AI space as easy as possible! It can be used in a variety of ways for example for ticket or e-mail classification.
Additional Information
Additional Information
Dependencies
UiPath.ML.Services.Activities >= 1.1.3 Newtonsoft.Json >= 12.0.3
Technical
Version
1.0Updated
February 4, 2021Works with
AI Center: 20.10 - 22.10
Certification
Gold Certified
Tags
Support
Email: support@roboyo.de
Mon-Fri 9 AM to 6 PM CET; Response time: 1 business day (24 hours); Resolution time: 10 days
Resources