I've had two client situations where the classifying keywords by their sentiment in the WDS tooling would be relevant. One was for an ES Learn where the client wanted to see the keywords with negative sentiment in Amazon reviews and although we are able to show that in the API we are not able to show that in the Tooling UI. When I started in October, I created a lab and the keyword sentiment feature was included in the tooling, but ever since that has been changed so I've had to adjust the lab.
Second client project was a Watson Business Solution Voice of the Customer engagement where the client wanted to see the keywords associated with negative sentiment in the WDS tooling, but we were not able to show that since we were only using WDS through the tooling. In these short engagements, it is important to show the client the value of WDS to get them to buy in further. We'd like to move the WBS VOC with our client into production, but they need to be excited and see the value of picking up positive keywords and negative keywords. When just showing sentiment generally and relating it to WKS entities we have created, there is not as much value. The client wants to see sentiment related to the keyword and not the overall file. It would be helpful if you guys could add back the enriched_review_text.keywords.sentiment.label or score. I understand it is available in the API and that is good to know, but it would be nice to showcase this feature neatly in the UI as it used to be. Thanks for your consideration!
Why is it useful?
|Who would benefit from this IDEA?||As a customer I would like to see the sentiment of specific keywords in an easy to view user interface|
How should it work?