Metadata Driven Data Science
Regular talk, 1:30 - 2:30 PM
Many data science teams face unprecedented demands for data products and actionable intelligence. This presentation will showcase the work that the Oregon EMS Program has done to build workflows using metadata to drive data extraction cleaning analysis and visualization. This approach allows us to get upstream of the many projects that we develop and/or update every year/quarter/month to make changes in only one place; our data dictionary. Using a modular approach functions and scripts are built to pull in information about data elements values relationships and metrics from metadata repositories and allow us to produce more consistent validated data products more quickly.
The presentation is designed to be accessible to a general audience regardless of programming/coding experience. All examples given will be using our EMS dataset and the R programming language and R Studio IDE but a similar approach could be implemented using any database or in other languages such as Python SAS etc.
 |
Pronouns: PRONOUNS HERELOCATION HEREBIO HERE |