In response to the Data Treasure Hunting challenge, the NYSpaceTag team built a tagging system that extracts natural keywords from titles and descriptions. It allows users to explore concepts, see related concepts, and drill down directly into the data.
The system can be easily replicated to process new datasets as they are added. The results are stored in a web-accessible database, which makes the enriched data easy to use for any app.
After the team successfully deployed the process for extracting core concepts, they needed to come up with a simple way for users to query the datasets. They developed an ingenious method, which allows a “fuzzy search” on the extracted concepts, and returns not only the most common key word but also a collection of related concepts. The visual graph based search makes it easy to search datasets, understand the connected concepts, and access the data, directly from the source.