Mapping Coordinates and Creating Plot Maps

Introduction

In another post, we used the widely available “chicago_crime” dataset to predict the classification of different crimes.

The data can be found here.

Here, we will try to map visualize select aspects of the data using a map of the city to gain a better understanding of Chicago’s overall crime picture.

# Filter the data 
murders <- dt3[primarytype == "HOMICIDE" & year == 2017]

# Extract the locations
locs <- data.table(table(murders$longitude, murders$latitude))

# Filter out the 0's
locs <- locs[N > 0]

# Name the columns
setnames(locs, c("V1", "V2"), c("longitude", "latitude"))

# Convert to numeric
locs$longitude <- as.numeric(locs$longitude)
locs$latitude <- as.numeric(locs$latitude)

# Get a map of Chicago
chi <- get_map(location = "Chicago", zoom = 11)

## Map from URL : http://maps.googleapis.com/maps/api/staticmap?center=Chicago&zoom=11&size=640x640&scale=2&maptype=terrain&language=en-EN&sensor=false

## Information from URL : http://maps.googleapis.com/maps/api/geocode/json?address=Chicago&sensor=false

# Plot points
ggmap(chi) + geom_point(data = locs, aes(x = longitude, y = latitude, color = N, size = N))

Full code

library(knitr)
library(data.table)
library(maps)
library(ggmap)
library(mapproj)
library(lubridate)

# Set the root directory for RMarkdown
opts_knit$set(root.dir = "/PATH")

set.seed(100)

options(scipen = 3)
knitr::opts_chunk$set(echo = TRUE)

# Use "fread" to load the data and create a data.table
dt <- setDT(fread("chicago_crime.csv"))

# Remove the unwanted characters
names(dt) <- gsub(" ", "", names(dt))

# Lower-casing
setnames(dt, names(dt), tolower(names(dt)))

# Fix the dates
date_convert <- as.Date(strptime(dt$date, "%m/%d/%Y %H:%M:%S"))
dt$date <- date_convert

# Create columns for month, day, and year
dt[, c("month", "day") := c(list(months(date_convert)), list(weekdays(date_convert)))]

# Convert factor columns
factors <- c("block", "iucr", "primarytype", "description", "locationdescription", "arrest", "domestic", "ward", "communityarea", "fbicode", "year", "month", "day")

dt2 <- dt[, (factors) := lapply(.SD, as.factor), .SDcols = factors]

# Combine non-criminal offenses to reduce number of levels of "primarytype" 
levels(dt2$primarytype)[c(20, 21, 22)] <- "NON-CRIMINAL"

# Remove non-criminal offenses from the data
dt3 <- dt2[dt2$primarytype != "NON-CRIMINAL",]

dt3$primarytype <- dt3$primarytype[, drop = TRUE]

# Filter the data 
murders <- dt3[primarytype == "HOMICIDE" & year == 2017]

# Extract the locations
locs <- data.table(table(murders$longitude, murders$latitude))

# Filter out the 0's
locs <- locs[N > 0]

# Name the columns
setnames(locs, c("V1", "V2"), c("longitude", "latitude"))

# Convert to numeric
locs$longitude <- as.numeric(locs$longitude)
locs$latitude <- as.numeric(locs$latitude)

# Get a map of Chicago
chi <- get_map(location = "Chicago", zoom = 11)

# Plot points
ggmap(chi) + geom_point(data = locs, aes(x = longitude, y = latitude, color = N, size = N))