GBAT.process_addresses: Geocode a large data frame of NYC addresses where there are...
In gmculp/rGBAT16AB: GeoSupport for R

Description Usage Arguments Details Value Examples

The GBAT.process_addresses function is a more efficient way to geocode a large data frame of NYC addresses where there are duplicates. It aggregates, geocodes and then merges the geocoded results back to the original data frame.

1 2	GBAT.process_addresses(in_df, add_col_name, third_col_name, source_cols, geocode_cols, add_type = "boro_code", return_type = "all")

`in_df`	a data frame containing NYC address
`add_col_name`	the column name of addresses where addresses are in the format of house number and street name (e.g., 123 Smith Street)
`third_col_name`	the column name of the boro_code or zip_code
`source_cols`	vector of column names from the input data frame to be returned with geocoder results
`geocode_cols`	vector of column names generated by the geocoder to be returned with geocoder results
`add_type`	either boro_code or zip_code
`return_type`	specify if all records returned or only passes (those that have GRC of 00 or 01)

The New York Department of City Planning's Geosupport Desktop geocoding software must be installed prior to installing the rGBAT16AB package.

The GBAT.process_addresses function can return almost 200 columns which are described in the Geosupport Desktop User Guide

# create a data frame of addresses
ADDR <- c("448 Lafayette Ave","727 Manhattan Ave","31-00 47th Ave","308 E 78th S","10 Morton St")
ZIP_CODE <- c("11205","11222","11101","10075","10014")
BORO_CODE <- c(3,3,4,1,1)
u_id <- 1:length(ADDR)
df = data.frame(u_id, ADDR, ZIP_CODE, BORO_CODE)

source_cols <- c('u_id')

#preview potential geocode return columns
paste(GC_output$GC_function,GC_output$GC_colname,sep="_")

geocode_cols <- c('F1E_BIN','F1E_BBL','F1E_Longitude','F1E_Latitude')

#geocode data frame
df_gc <- GBAT.process_addresses(df, "ADDR", 'ZIP_CODE', source_cols, geocode_cols, "zip_code","all")