Abstract

City of Portland's Property Data Cleanup
Track: Big Data
Authors: Jake Brown

Portland has created a Master Address Repository integrating address points, building and tax lot polygons. Because this property data was derived from multiple sources, there is a large amount of error in the dataset. A series of tests have been created to find errors and create a score for a feature’s spatial accuracy and attribute content. Python is used to query commercial web services and provide additional indications of error. These scores are used to direct editors cleaning up the data.