README.md

Use https://github.com/geomarker-io/parcel instead

hamilton

Offline parcel-based geocoding for addresses in Hamilton County, Ohio, USA

This package is designed to geocode address strings using an offline copy of the Cincinnati Area Geographic Information System (CAGIS) master address/parcel files. This drastically improves geocoding accuracy, but is only available for addresses that have a zipcode beginning with 450, 451, or 452, i.e. the Greater Cincinnati Area.

Its major functions include:

• parse_address()
  • parse an address string into individual components
  • converts all letters to lower case
  • replaces consecutive whitespace with one whitespace character
  • removes non-alphanumeric characters
  • takes an address string and returns a tibble of address components
• geocodeCAGIS()
  • parcel-based geocoding based on included CAGIS address database
  • fuzzy text matching on street will return the best match and its similarity score
  • return CAGIS parcel id to link with other extant databases including Hamilton County Auditor

Installation

Install the package by running the following in R:

remotes::install_github('cole-brokamp/hamilton')

The package relies on python to parse the address, which means that both python and the usaddress python library must be installed (pip install usaddress). Alternative installation methods are available depending on your operating system; seek instructions online.

Usage

Geocoding

Geocode a single address string using geocodeCAGIS(). This will automatically take care of parsing the address string and will return the latitude and longitude, the CAGIS parcel identifier, and optionally, some additional diagnostic information regarding the geocoding match process.

geocodeCAGIS('3333 Burnet Ave, Cincinnati, OH 45229')
#> # A tibble: 1 x 5
#>     lat   lon PARCELID     score match               
#>   <dbl> <dbl> <chr>        <dbl> <chr>               
#> 1  39.1 -84.5 010400020052     1 3333 burnet av 45229

Address String Formatting

• An address must be submitted as a single string
• The city and state are optional and not used for geocoding.
• Street number, street name, and zip code must be in the address string.
• Street numbers must be numeric (i.e. “18”, not “Eighteen”).

Batch Geocoding

Geocoding an address string requires a significant amount of computational time, so it may be useful to memoise results, provide a progress bar, implement graceful error handling, and/or use parallel processing with CB::mappp():

addrs <- c('3333 Burnet Ave, Cincinnati, OH 45229',
           '222 E Central Pkwy, Cincinnati, OH 45202',
           '1600 Pennsylvania Ave NW Washington, DC 20500')

CB::mappp(addrs, geocodeCAGIS)
#> 
...  :what (  0%)   [ ETA:  ?s | Elapsed:  0s ]
...  processing 1 of 3 ( 33%)   [ ETA:  0s | Elapsed:  0s ]
...  processing 2 of 3 ( 67%)   [ ETA:  0s | Elapsed:  1s ]
...  processing 3 of 3 (100%)   [ ETA:  0s | Elapsed:  1s ]
#> warning: zip code does not begin with 450, 451, or 452; returning NA
#> [[1]]
#> # A tibble: 1 x 5
#>     lat   lon PARCELID     score match               
#>   <dbl> <dbl> <chr>        <dbl> <chr>               
#> 1  39.1 -84.5 010400020052     1 3333 burnet av 45229
#> 
#> [[2]]
#> # A tibble: 1 x 5
#>     lat   lon PARCELID     score match                   
#>   <dbl> <dbl> <chr>        <dbl> <chr>                   
#> 1  39.1 -84.5 007500040247     0 222 central pkwy 45202 e
#> 
#> [[3]]
#> [1] NA