This idea has been merged into another idea. To comment or vote on this idea, please visit WDS-I-72 Ability re-process documents without re-ingestion.
In case of applying WKS's custom model (or using custom tokenization function available only in Japanese language), there are the following issues. In IBM Disocovery specifications, we can not re-index documents that have been submitted to existing collections, so in order to update WKS's model (or tokenized dictionary) and use it for Discovery's collection, It takes much time and effort because we need to do tasks. 1. Create a new collection 2-1. Set custom model to configuration (Or set custom tokenized dictionary with API) 3. Populated input document into new collection just like existing collection We are developing and selling solutions using IBM Discovery as With Watson Premier Member. In order to make this work more efficient, we are considering functionalization to solutions, but we would like to wait for implementation if there is a productization plan.
Why is it useful?
|Who would benefit from this IDEA?||Using WKS Custom model with WDS , and all WDS Customer in Japan.|
How should it work?