1 Map of london businesses from London Datastore

Reference URLs

http://www.sthda.com/english/wiki/computing-and-adding-new-variables-to-a-data-frame-in-r
https://data.london.gov.uk/dataset/strategic-industrial-location-points-london-plan-consultation-2009
https://data.london.gov.uk/dataset/strategic-industrial-location-points-london-plan-consolidated-alterations-2004
https://data.london.gov.uk/dataset/town-centre-locations-2015
https://data.london.gov.uk/dataset/statistical-gis-boundary-files-london
https://data.cdrc.ac.uk/dataset/tlrn
https://data.london.gov.uk/dataset/directory-of-london-businesses

libraries

library(dplyr)
library(readr)
library(maptools)
library(RColorBrewer)
library(classInt)
#library(OpenStreetMap)
library(sp)
library(rgeos)
library(tmap)
library(tmaptools)
library(sf)
library(rgdal)
library(geojsonio)
library(ggplot2)
library(spData)
library(tidyverse)

1.1 prepare dataset

Read from RDS server

bil <- read_csv("/Volumes/ritd-ag-project-rd00lq-jamfe87/GIS_Analysis/dataRaw/businesses-in-london.csv")

Create variable ‘division’ from variable ‘SICCode.SicText_1’ substract chr 1 to 4 (first 4 characters)

bil <- mutate(bil, division = substr(SICCode.SicText_1, 1, 4))

Change chr to num for ‘division’ variable

bil$division <- as.numeric(as.character(bil$division))

Convert numeric to factor to create Division labels

bil$divname <- cut(bil$division, 
                  breaks = c(-Inf, 999, 1499, 1799, 1999, 3999, 4999, 5199, 5999, 6799, 8999, 9729, 9999),
                  labels = c("Agriculture, Forestry and Fishing", "Mining", "Construction", "not used", "Manufacturing","Transportation, Communications, Electric, Gas and Sanitary service", "Wholesale Trade", "Retail Trade", "Finance, Insurance and Real Estate", "Services", "Public Administration", "Nonclassifiable"), right = TRUE)

Count number of variables in ‘divname’

bil$divname <- as.character(bil$divname)
# length(unique(bil[["divname"]])) # [1] 13 - Not necessary, see next results from next function
library(plyr)
library(knitr)
sumt <- count(bil, vars = "divname")
kable(sumt, caption = "Division Frequency")
Division Frequency
divname freq
Agriculture, Forestry and Fishing 4358
Construction 1566
Finance, Insurance and Real Estate 139254
Manufacturing 22320
Mining 9410
Nonclassifiable 79513
not used 2499
Public Administration 90279
Retail Trade 86037
Services 459815
Transportation, Communications, Electric, Gas and Sanitary service 195103
Wholesale Trade 2662
NA 37658

1.2 generate quick map

Convert to dataframe - to SP object - to SF object

bildf <-as.data.frame(bil)

Get long and lat from your data.frame. Make sure that the order is in lon/lat.

latlon <- bildf[,c('long','lat')]
bilsp <- SpatialPointsDataFrame(coords = latlon, data = bil,
                               proj4string = CRS("+proj=longlat +datum=WGS84 +ellps=WGS84 +towgs84=0,0,0"))

SP to SF

bilsf <- st_as_sf(bilsp)

Subset by ‘Manufacturing’ . Plotting ‘bil’ won’t work quickly because N = 1,130,474 obs.

man <- subset(bilsp, divname == "Manufacturing")

Plot quick map “Manufacturing”

qtm(man)

Style map ‘Manufacturing’

tmap_mode("plot") # or "plot" or "view"
tm_shape(man) + 
  tm_symbols(size=0.1,
             shape = 20,
             col = "magenta", 
             alpha = .1,
             border.lwd = NA
  )

1.3 interactive map manufacturing

# tmap_mode("view") # or "plot" or "view"
# tm_shape(man) + 
#   tm_symbols(size=0.001,
#              shape = 20,
#              col = "magenta", 
#              alpha = .1,
#              border.lwd = NA
#   )