Search engines for data terms
The challenge
When building the new Data Warehouse, we have to deal with various data sources to extract business/data terms which will be use later for business. However, we face some fundamental but challenging questions such as:
- How could I know that a term/multiple terms have been developed in the new DWH?
- How could I find data terms effectively as we face problems of Natural Language Processing with Vietnamese language?
Solution
To solve such problems, we built a search engine using Natural Language Processing
with Shiny
in IBM Watson platform. The app help us to have:
- Simple but modern interface for business users to find data available in the new DWH
- Effectve search engine which can identify multiple terms with propensity of accuracy
Thanks to the app, it has saved hundreds hours for users/project to search and communicate to each other.