A Novel Method of Location Based Listing of Residential Addresses

We have entered into a new era of the geo-information revolution. A vast amount of data/information that we generate is georeferenced and available publicly. This article demonstrates the application of reverse geocoding couple with web-mining for developing a georeferenced list of residential addresses using these datasets. The proposed methodology is likely to extend the scope of spatial sampling for social, demographic and health surveys (SDHS). Spatial sampling for SDHS involves sampling locations/areas first and then sampling respondents at/around these locations. Four different methods of spatial sampling were employed to draw four sets of samples in the Chicago Metropolitan Statistical Area (MSA). A list of residential addresses at/around each sample site was developed with the aid of reverse geocoding and web-mining using publicly available datasets; a residential address was selected randomly from this list. The sampled residential addresses were compared with the US Postal Service’s address database; 98.7% of the addresses that we generated using reverse geocoding and web-mining were valid residential addresses. Results are important, and suggest that the availability of geo-referenced data provides us with an unprecedented opportunity to develop a list of residential addresses in a cost-effective manner, and augment the scope of spatial sampling for social surveys