A s part of a POC, we have built a translation tool for a financial industry client. The tools uses the latest Watson translation engines, and the primary model is Japanese to English translation. In order to improve the translation quality, we added in a feature for the customer to submit wrongly translated terms pairs, which are stored in a Cloudant database. This data will be used to modify and build a customized Japanese model for the client. In order to visualize and analyze the high frequency mistranslated pairs, we are attempting to build a dashboard that visualizes all the user feedback data from Cloudant. Right now, importing the data from Cloudant to build this dashboard is a limitation and not supported. We plan to present a full solution design to the client, but will have adjust scope due to DB2 and postgress being the only DBs supported at the time.
Why is it useful?
|Who would benefit from this IDEA?||Developers, watson language annotators , and data anlaysts, and the client, would be able to benefit from Cloudant support , so they can visualize their data in a dsx dashboard|
How should it work?