DSC-611-Z1 Data Visualization (Module 6 - Dot Maps and Google API)

Master in Data Science at Utica College

https://programs.online.utica.edu/programs/masters-data-science

Clearing R Studio Memory Usage

gc()

##           used (Mb) gc trigger (Mb) max used (Mb)
## Ncells  542109 29.0    1238217 66.2   621331 33.2
## Vcells 1019753  7.8    8388608 64.0  1600889 12.3

rm(list = ls())

Time Counter Start

start_time <- Sys.time()

Include the knitr package for integration of R code into Markdown

knitr::opts_chunk$set(echo = TRUE)

All the libraries used in this code

library(easypackages)
libraries("data.table","devtools","dplyr","ggplot2","ggmap","ggrepel","ggsn","lubridate","readr","tidyr","tidyverse","sf")

Download the main crime incident dataset

POI= read.csv('Point_Of_Interest.csv', header=T)

Descriptive statistics before data processing

Dimension of data frame

dim(POI)

## [1] 19010    17

Structure of data frame

str(POI)

## 'data.frame':    19010 obs. of  17 variables:
##  $ SEGMENTID : int  69163 34217 34220 43117 9008897 78137 9004671 17826 67394 34220 ...
##  $ LON       : num  -73.9 -74 -74 -73.9 -73.8 ...
##  $ LAT       : num  40.8 40.8 40.8 40.7 40.9 ...
##  $ COMPLEXID : int  0 0 0 0 515 0 2245 0 0 0 ...
##  $ SAFTYPE   : Factor w/ 4 levels "","G","N","X": 3 1 1 1 1 1 4 3 1 1 ...
##  $ SOS       : int  2 1 1 2 1 NA 2 2 2 1 ...
##  $ PLACEID   : int  12171 13932 2410 3247 8230 5077 7129 13182 15765 5444 ...
##  $ FACI_DOM  : int  10 3 4 1 2 2 3 7 1 3 ...
##  $ BIN       : int  1074333 1024838 1024839 3047371 2093858 2009619 3327986 3332515 4436782 1024843 ...
##  $ BOROUGH   : int  1 1 1 3 2 2 3 3 4 1 ...
##  $ CREATED   : Factor w/ 1147 levels "01/02/2013 12:00:00 AM +0000",..: 447 447 447 447 447 447 447 447 447 447 ...
##  $ MODIFIED  : Factor w/ 1438 levels "01/02/2013 12:00:00 AM +0000",..: 1303 1207 1207 1207 1207 1305 1011 1379 1207 1207 ...
##  $ FACILITY_T: int  2 3 7 9 1 5 1 8 9 3 ...
##  $ SOURCE    : Factor w/ 11 levels "DCP","DOE","DoITT",..: 2 6 9 9 3 3 1 1 5 6 ...
##  $ B7SC      : int  10078501 NA NA NA NA NA 32055002 33361901 NA NA ...
##  $ PRI_ADD   : int  5128938 1023588 1023591 3058824 2101927 2017249 3094096 5125956 0 1023594 ...
##  $ NAME      : Factor w/ 18875 levels "0 BOND STREET",..: 13632 5082 500 6650 3575 9090 1707 4702 16199 16527 ...

Summary statistics of data frame

summary(POI)

##    SEGMENTID            LON              LAT          COMPLEXID      SAFTYPE  
##  Min.   :     10   Min.   :-74.27   Min.   :40.45   Min.   :   0.0    :11219  
##  1st Qu.:  39984   1st Qu.:-73.98   1st Qu.:40.67   1st Qu.:   0.0   G:  662  
##  Median : 109339   Median :-73.94   Median :40.73   Median :   0.0   N: 2847  
##  Mean   :1623955   Mean   :-73.94   Mean   :40.73   Mean   : 260.6   X: 4282  
##  3rd Qu.: 268553   3rd Qu.:-73.89   3rd Qu.:40.80   3rd Qu.: 218.0            
##  Max.   :9024271   Max.   :-73.71   Max.   :40.95   Max.   :4300.0            
##                                                                               
##       SOS           PLACEID           FACI_DOM           BIN         
##  Min.   :1.000   Min.   :      1   Min.   : 1.000   Min.   :      0  
##  1st Qu.:1.000   1st Qu.:   7054   1st Qu.: 2.000   1st Qu.:1004331  
##  Median :2.000   Median :  13796   Median : 3.000   Median :2010983  
##  Mean   :1.511   Mean   : 325108   Mean   : 4.312   Mean   :2030343  
##  3rd Qu.:2.000   3rd Qu.:1010060   3rd Qu.: 6.000   3rd Qu.:3326732  
##  Max.   :2.000   Max.   :1031244   Max.   :18.000   Max.   :5169276  
##  NA's   :2081                                                        
##     BOROUGH                              CREATED     
##  Min.   :1.000   05/14/2009 12:00:00 AM +0000:12800  
##  1st Qu.:1.000   09/15/2008 12:00:00 AM +0000:  200  
##  Median :3.000   08/17/2011 12:00:00 AM +0000:  121  
##  Mean   :2.564   08/16/2011 12:00:00 AM +0000:   93  
##  3rd Qu.:4.000   09/03/2010 12:00:00 AM +0000:   62  
##  Max.   :5.000   09/20/2012 12:00:00 AM +0000:   53  
##  NA's   :221     (Other)                     : 5681  
##                          MODIFIED       FACILITY_T         SOURCE    
##  11/30/1899 12:00:00 AM +0000: 2139   Min.   : 1.000   OTHER  :5056  
##  11/24/2014 12:00:00 AM +0000: 1543   1st Qu.: 2.000   DoITT  :3392  
##  12/09/2010 12:00:00 AM +0000:  148   Median : 4.000   DCP    :2753  
##  11/13/2017 12:00:00 AM +0000:  100   Mean   : 4.784   DOE    :2017  
##  12/11/2012 12:00:00 AM +0000:   69   3rd Qu.: 7.000   NYPD   :1993  
##  01/29/2013 12:00:00 AM +0000:   68   Max.   :13.000   EMS    :1441  
##  (Other)                     :14943                    (Other):2358  
##       B7SC             PRI_ADD                         NAME      
##  Min.   :10000201   Min.   :       0   DEVRY COLLEGE     :    3  
##  1st Qu.:13289506   1st Qu.:       0   HOLY ROSARY CHURCH:    3  
##  Median :30013851   Median : 1031109   HOLY ROSARY SCHOOL:    3  
##  Mean   :27945586   Mean   : 1810248   PS 12             :    3  
##  3rd Qu.:38334504   3rd Qu.: 3109959   PS 146            :    3  
##  Max.   :57019202   Max.   :10173608   PS 15             :    3  
##  NA's   :11270                         (Other)           :18992

Subsetting Data Set

POI = POI[which(POI$BOROUGH>0),]

Add Variable BOROUGHNAME

numbers=sort(unique(POI$BOROUGH))
boroughs=c("Manhattan","Bronx","Brooklyn","Queens","Staten Island")
names(boroughs)=numbers
POI$BOROUGHNAME=boroughs[POI$BOROUGH]

Set API Key ~ put key between the quotations marks

ggmap::register_google(key ="AIzaSyD9jYJNqrIoumMz9pIi6MHshWi20IYilUw")

Define longitude and latitude of the center of your map (currently Williamsburg, Brooklyn, NYC)

center <- c(lon = -73.955, lat = 40.715)

Load the basemap

#experiment with different maptypes ("roadmap", "satellite", "terrain", "hybrid"), and zoom levels (3 - 21)
p1 <- ggmap(get_googlemap(center, maptype = "satellite", zoom=13))
p2 <- ggmap(get_googlemap(center, maptype = "roadmap", zoom=13))

Plot the POI points upon the basemap

scalebar1 = function(x,y,w,n,d, units){
  # x,y = Lower left coordinate of bar
  # w = Thickness of bar
  # n = Number of divisions on bar
  # d = Distance along each division (longitude degree)

  bar = data.frame( 
    xmin = seq(0.0, n*d, by=d) + x,
    xmax = seq(0.0, n*d, by=d) + x + d,
    ymin = y,
    ymax = y+w,
    z = rep(c(1,0),n)[1:(n+1)],
    fill.col = rep(c("blue","pink"),n)[1:(n+1)])

  labs = data.frame(
    xlab = c(seq(0.0, (n+1)*d, by=d) + x, x), 
    ylab = c(rep(y-w*1.5, n+2), y-3*w),
    text = c(as.character(  seq(0.0, ((n+1)*d)*55.051, by= round(d*55.051,1) )), units)
    )
  list(bar, labs)
}

sb1 = scalebar1(-73.97, 40.68, 0.001, 2, 0.015, "Miles")


scalebar2 = function(x,y,w,n,d, units){
  # x,y = Lower left coordinate of bar
  # w = Thickness of bar
  # n = Number of divisions on bar
  # d = Distance along each division (longitude degree)

  bar = data.frame( 
    xmin = seq(0.0, n*d, by=d) + x,
    xmax = seq(0.0, n*d, by=d) + x + d,
    ymin = y,
    ymax = y+w,
    z = rep(c(1,0),n)[1:(n+1)],
    fill.col = rep(c("azure4","white"),n)[1:(n+1)])

  labs = data.frame(
    xlab = c(seq(0.0, (n+1)*d, by=d) + x, x), 
    ylab = c(rep(y-w*1.5, n+2), y-3*w),
    text = c(as.character(  seq(0.0, ((n+1)*d)*55.051, by= round(d*55.051,1) )), units)
    )
  list(bar, labs)
}

sb2 = scalebar2(-73.97, 40.68, 0.001, 2, 0.015, "Miles")


#experiment with different point sizes and colors
p1a <- p1 + geom_point(aes(x = LON, y = LAT, col=BOROUGHNAME), data = POI, size = 0.6,) +
       scale_colour_manual(name="Borough Name", 
                           values = c("Bronx"="black", "Brooklyn"="red",
                                      "Queens"="green", "Manhattan"="blue",
                                      "Staten Island"="orange")
                           ) + 
       geom_segment(arrow=arrow(length=unit(4,"mm")),aes(x=-73.989,xend=-73.989,y=40.685,yend=40.695),color="pink", size = 2) +
       annotate(x=-73.989, y=40.682, label="N", color="pink", geom="text", size=10, fontface = "bold") +
       geom_rect(data=sb1[[1]], aes(xmin=xmin, xmax=xmax, ymin=ymin, ymax=ymax, fill=z), inherit.aes=F, show.legend = F, color = "green", fill = sb1[[1]]$fill.col, size=0.5, alpha=5) +
       geom_text(data=sb1[[2]], aes(x=xlab, y=ylab, label=text), inherit.aes=F, show.legend = F, color = "green", size=3) +
       xlab("Longitude") + 
       ylab("Latitude") +
       ggtitle("Google Satellite Map - NYC Points of Interest") + 
       theme(plot.title = element_text(hjust = 0.5)) +
       theme(
             plot.title = element_text(colour = "black"),
             panel.border = element_rect(colour = "black", fill=NA, size=1.5)
            ) 
  
p1a

## Warning: Removed 14627 rows containing missing values (geom_point).

DSC-611-Z1 Data Visualization (Module 6 - Dot Maps and Google API)

By Henry J. Hu

Date: 10/30/2020

Master in Data Science at Utica College

Clearing R Studio Memory Usage

Time Counter Start

Include the knitr package for integration of R code into Markdown

All the libraries used in this code

Download the main crime incident dataset

Descriptive statistics before data processing

Dimension of data frame

Structure of data frame

Summary statistics of data frame

Subsetting Data Set

Add Variable BOROUGHNAME

Set API Key ~ put key between the quotations marks

Define longitude and latitude of the center of your map (currently Williamsburg, Brooklyn, NYC)

Load the basemap

Plot the POI points upon the basemap