hartford-police.Rmd

---
title: "Reported Crime Incidents in Hartford, CT"
author: "By Jasmine Dumas"
output: 
  flexdashboard::flex_dashboard:
    theme: yeti
    orientation: columns
    vertical_layout: fill
    social: [ "twitter", "facebook", "menu" ]
    source_code: embed
runtime: shiny
---
  
```{r setup, include=FALSE}
library(flexdashboard)
library(DT)  # dev version devtools::install_github("rstudio/DT")
library(leaflet)
library(readr)
library(lubridate)
library(ggplot2)
library(plotly)
library(shiny)
library(RCurl)
library(RSocrata)
```

Sidebar {.sidebar}
=====================================
  
Explore reported incidents of crime (with the exception of sexual assaults, which are excluded by statute) that occurred in the City of Hartford from 2005 to the present, minus the most recent ten days from the [Hartford Open Data Portal](https://data.hartford.gov/). 


```{r, shiny_in}
# shiny inputs defined here
hr()

selectInput("year", "Select Year: ", choices =c(2005, 2006, 2007, 
2008, 2009, 2010,
2011, 2012, 2013, 
2014, 2015, 2016, 2017) ,
selected = 2017)

hr()

```

Made for fun & interest by [Jasmine Dumas](http://jasdumas.github.io/).


Explore
=====================================

```{r, message=FALSE, warning=FALSE, include=FALSE}

# URL = "https://data.hartford.gov/api/views/889t-nwfu/rows.csv?accessType=DOWNLOAD"
# x <- getURL(URL, ssl.verifypeer = FALSE)
# out <- read_csv("https://data.hartford.gov/Public-Safety/Police-Incidents-01012005-to-Current/889t-nwfu")

#data_in <- read_csv("Police_Incidents_01012005_to_Current.csv") 

# get the data from data.hartford.gov as the API url
data_in <- read.socrata("https://data.hartford.gov/Public-Safety/Police-Incidents-01012005-to-Current/889t-nwfu")
##############################
# clean up lat and longitude #
##############################
# removes the parenthesis
s = gsub(".*\\((.*)\\).*", "\\1", data_in$geom)  
# extract and split each f the tuple latitude and longitude
lat_splat = sapply(s, function(x) strsplit(x, ",")[[1]][1])
lon_splat = sapply(s, function(x) strsplit(x, ",")[[1]][2])
# convert to data.frame
lat <- as.data.frame(lat_splat)
lon <- as.data.frame(lon_splat)
# convert to factor to numeric
options(digits=15)
lat2 <- sapply(lat, function(x) as.numeric(levels(x))[x])
lat2 <- as.data.frame(lat2)
lon2 <- sapply(lon, function(x) as.numeric(levels(x))[x])
lon2 <- as.data.frame(lon2)
# add to main data frame
data_in$lat <- lat2
data_in$lon <- lon2 
#################
# clean up date #
#################
class(data_in$Date)
data_in$Date <- mdy(data_in$Date)
data_in$year <- year(data_in$Date)
#####################
# clean up category #
#####################
## grep out leader symbol
data_in$UCR_1_Category <- gsub("[0-9]","", data_in$UCR_1_Category)
data_in$UCR_1_Category <- gsub("[[:punct:]]","", data_in$UCR_1_Category)

#######################################
# subset data by date for explore tab #
#######################################
## so many points!! - use sample for now
data_in <- as.data.frame(data_in)

data_sample <- reactive({
data_sample <- data_in[which(data_in$year == input$year), ] 
data_sample
})

```


### Map

```{r, map}

renderLeaflet({
m <- leaflet(data=data_sample()) %>%
setView(lng = -72.685097, lat = 41.763710, zoom = 11) %>%
addProviderTiles("Stamen.Toner") %>% 
addMarkers(lng=~lon$lon_splat, lat=~lat$lat_splat, 
popup = ~as.character(UCR_2_Description), 
clusterOptions = markerClusterOptions())
m  
})

```


### Data Table

```{r, dt}

DT::renderDataTable({

DT::datatable(data_sample(), options = list(bPaginate = TRUE,
pageLength = 15
))
})
```



Data Visualization {.storyboard}
=====================================

### **Crime by Neighborhood:** Which neighboorhoods have had the most reported incidents?

```{r, viz_display2}
category <- dplyr::count(data_in, UCR_1_Category)

neighborhood <- dplyr::count(data_in, UCR_1_Category, Neighborhood)

g2 <- ggplot(neighborhood, aes(x=reorder(Neighborhood, n), y = n, fill=factor(UCR_1_Category))) +
      geom_bar(stat="identity") +
      theme(axis.text.x  =element_blank()) +
      theme(legend.title=element_blank()) +
      theme_bw() +
      coord_flip() +
      labs(title="Overall Count of Incidents by Crime Type & Neighborhood", 
      x = "", y = "Count")

ggplotly(g2)

```


###  **Decrease in Reported Crimes:** Has the amount of reported crime incidents been decreasing over the years?

```{r}
safe <- dplyr::count(data_in, year)

p1 <- ggplot(safe, aes(x = year, y = n)) +
             geom_line(color = "dodgerblue") +
             geom_point(color = "dodgerblue") +
             theme_bw() +
             labs(title = "Annual Amount of Reported Incidents", x = "", y = "Count")

ggplotly(p1)

```


### **Decrease in Violent Crimes:** Can lower amounts of reported crime incidents be related to fewer violent crimes?

```{r}
violent <- dplyr::count(data_in, UCR_1_Category, year)

really_violent <- c("  AGGRAVATED ASSAULT", "  ROBBERY", "  HOMICIDE", "  LARCENY")

violent2 <- violent[violent$UCR_1_Category %in% really_violent,]

vi <-  ggplot(data=violent2, aes(x=year, y=n, group=UCR_1_Category, color=UCR_1_Category)) +
    geom_line() +
    geom_point() +
    theme(legend.position="none") +
    theme_bw() +
    labs(title = "Annual Amount of Reported Violent Crimes", x = "", y = "Count")

ggplotly(vi)
```

### **Guns in Hartford:** As a National issue about gun control and safety, [Conneticut has enacted the toughest restrictions on purchasing weapons and ammunition](https://en.wikipedia.org/wiki/Gun_laws_in_Connecticut) since 2012.

```{r}
gun_related <- c("  SHOOTING", "  SHOTS FIRED", "  SHOTS FIRED CONFIRMED", "  SHOTS FIRED UNCONFIRMED")

gun_df <- dplyr::count(data_in, UCR_1_Category, year)
gun_df2 <- gun_df[gun_df$UCR_1_Category %in% gun_related, ]

gu <- ggplot(gun_df2, aes(x = UCR_1_Category)) +
             geom_bar(fill = "#536267") +
             theme_bw() +
             labs(title = "Amount of Gun-Related Reported Crimes", x = "", y = "Count")

ggplotly(gu)

```


### **The Drug Epidemic:** [Connecticut](http://overdose.trendct.org/) continues to see a sharp rise in heroin and other opioid-related deaths.

```{r}
drugs <- c("  DRUG OFFENSES")

drugs_df <- dplyr::count(data_in, UCR_1_Category, Neighborhood) 

drugs_df <- drugs_df[drugs_df$UCR_1_Category %in% drugs, ]

##
d <- ggplot(drugs_df, aes(x=reorder(Neighborhood, n), y = n)) +
     geom_bar(stat = "identity", fill = "#6699CC") +
     theme_bw() +
     coord_flip() +
     labs(title = "Amount of Drug-Related Reported Crimes", y = "", x = "") 

ggplotly(d)

```



Details
=====================================


* This dataset reflects reported incidents of crime (with the exception of sexual assaults, which are excluded by statute) that occurred in the City of Hartford from 2005 to the present, minus the most recent ten days. Data is extracted from the City of Hartford Police Department's CrimeView database on a daily basis. Should you have questions about this dataset, you may contact the Crime Analysis Division of the Hartford Police Department at 860.757.4020 or policechief@Hartford.gov. 

* Full Disclaimer available at [https://data.hartford.gov/Public-Safety/Police-Incidents-01012005-to-Current/889t-nwfu](https://data.hartford.gov/Public-Safety/Police-Incidents-01012005-to-Current/889t-nwfu)

* [Socrata Hartford Police Codes (Download)](https://data.hartford.gov/api/views/889t-nwfu/files/BOGvjBIm4--AIkVZZIBXBljNj-PrylsI5BxLEEQ4Lms?download=true&filename=Socrata%20HPD%20UCR%20Codes.pdf)

* Project inspiration: [Real-time Dallas Police Calls](https://trestletech.shinyapps.io/dallas-police/)