Handle rating scale conversions between Documents rated using API and documents rated using Static relevancy training.
We are an OEM partner of IBM and have built a Chatbot integration with Watson Discovery search.
The end Users of Chatbot rate individual documents that turn up in search from Chatbot using a rating scale of 1 Star to 5 Star.
We convert the Star rating as 2, 4, 6, 8, 10 and then use the Discovery create training example API https://cloud.ibm.com/apidocs/discovery?code=java#createtrainingexample for adding a Discovery Training example based on the Chatbot end User rating.
The rating 2, 4, 6. 8, 10 is passed as relevance parameter of the Training Example passed to the Discovery API.
Besides rating using API as above, our Customer's want to also use static relevancy training.
However once Discovery documents are rated based on Chatbot end User rating as above, in Discovery UI when looking at queries to be rated such documents show the error "Rated with incompatible scale".
We located a IBM document https://www.ibm.com/support/pages/why-does-rated-incompatible-scale-error-occur which mentions that if we plan to rate the document with both API as well as Discovery tooling, then the documents must be rated as 0 (Not relevant) or 10 (Relevant). This means end User can not provide graded rating to provide relative usefulness of a document.
The above looks like an artificial restriction. Any rating number between 0 to 10 or in general any arbitrary rating scale should be compatible in terms of rating scale for documents rated by API and documents rated by Discovery tooling. Discovery should be able to handle rating scale conversion between documents rated using API and documents rated using Static relevancy training.
This will help Customers leverage rating from API as well as Static relevancy training without any restriction on the rating scale.
|Who would benefit from this IDEA?||Discovery customers who want to rate documents with arbitrary scale using API as well as who want to also use Static relevancy training|
|IBM's success depends on gathering feedback from customers like yourself. Aha Ideas Portal is the third party tool through which IBM Offering Managers gather feedback from customers such as yourself.|
|IBM is a global organization with business processes, management structures, technical systems and service provider networks that cross borders. As such, the information collected through Aha Ideas Portal (Customer Name, Customer Email Address) will be stored by them in the United States, and handled only as per IBM's instructions and policies. Your data (Name and Email Address) will NOT be shared with other IBM customers.|
|In order to safeguard your information in Aha, do not leave your workstation unattended while using this application, log off after using it, and print only if necessary. If you need to make a hardcopy, remember to pick up the print-out immediately, keep it under lock, and destroy it immediately when no longer needed.|
|NOTICE TO EU RESIDENTS: per EU Data Protection Policy, if you wish to remove your personal information from the IBM ideas portal, please login to the ideas portal using your previously registered information then change your email to "firstname.lastname@example.org" and first name to "anonymous" and last name to "anonymous". This will ensure that IBM will not send any emails to you about all idea submissions|