Project Proposal

Introduction:

This project is primarily focused in anticipating crime as per gender, age and race in the city of Cincinnati in order to create social security awareness among citizens as well as to target education materials.

Objective:

I will be using the City of Cincinnati Police Crime dataset to anticipate various crime as per gender, age and race. This will help in targeted educational materials across the city to create social security awareness as well as will help the citizens of Cincinnati to understand what kind of offense is more prevalent based on age and race.

Introduction to dataset:

I will be using City of Cincinnati Police Crime dataset which has approx. 358k rows and 40 columns. Out of these columns, my response variable will be “VICTIM_GENDER” and my predictor variables will be OFFENSE, VICTIM_AGE AND VICTIM_RACE. Since VICTIM_GENDER is a categorical variable, this will be the best target variable to analyze the data and get the outcome.

Packages Required:

To analyze this data, we will use the following R packages:

library(dplyr)
library("rio")

• DPLYR - It is a grammar of data manipulation, providing a consistent set of verbs that help you solve the most common data manipulation challenges:

mutate() adds new variables that are functions of existing variables select() picks variables based on their names. filter() picks cases based on their values. summarise() reduces multiple values down to a single summary. arrange() changes the ordering of the rows.

These all combine naturally with group_by() which allows you to perform any operation “by group”. dplyr is designed to abstract over how the data is stored.

• RIO - It implements format-specific S3 methods for each type of file that can be imported from or exported to. This happens via internal S3 generics, .import and .export. It is possible to write new methods like with any S3 generic (e.g., print). As an example, .import.rio_csv imports from a comma-separated values file.

Data Preparation:

The data set comes from the Cincinnati Police Crime Data.

Incidents are the records, of reported crimes, collated by an agency for management. Incidents are typically housed in a Records Management System (RMS) that stores agency-wide data about law enforcement operations.This dataset contains 40 columns and approx 358k rows. This dataset was created in November 15, 2017 but has been updated recently on April 4, 2019.

Below are the steps which have been followed for data importing and cleaning:

• Data Import - To work with the dataset, it needs to be loaded into R Studio. “read_csv” command is used to load the csv file into R Studio.

crime_data <- import("C:/Users/Ankeeta/Desktop/city_of_cincinnati_police_data_initiative_crime_incidents.csv")
glimpse(crime_data)

## Observations: 355,379
## Variables: 40
## $ INSTANCEID                     <chr> "D6D7D173-E416-4571-AF34-A767AC...
## $ INCIDENT_NO                    <chr> "159006170", "41103934", "21104...
## $ DATE_REPORTED                  <chr> "03/16/2015 05:19:00 PM +0000",...
## $ DATE_FROM                      <chr> "03/16/2015 03:02:00 PM +0000",...
## $ DATE_TO                        <chr> "03/16/2015 03:05:00 PM +0000",...
## $ CLSD                           <chr> "J--CLOSED", "J--CLOSED", "D--V...
## $ UCR                            <int> 802, 810, 1493, 401, 810, 550, ...
## $ DST                            <chr> "4", "4", "2", "5", "1", "2", "...
## $ BEAT                           <chr> "2", "3", "3", "4", "3", "3", "...
## $ OFFENSE                        <chr> "AGGRAVATED MENACING", "ASSAULT...
## $ LOCATION                       <chr> "48-PARKING LOT", "02-MULTI FAM...
## $ THEFT_CODE                     <chr> "", "", "", "", "", "", "", "",...
## $ FLOOR                          <chr> "", "", "", "", "", "2 - FIRST ...
## $ SIDE                           <chr> "", "", "", "", "", "1 - FRONT"...
## $ OPENING                        <chr> "", "", "", "", "", "1 - DOOR",...
## $ HATE_BIAS                      <chr> "N--NO BIAS/NOT APPLICABLE", "N...
## $ DAYOFWEEK                      <chr> "MONDAY", "THURSDAY", "SATURDAY...
## $ RPT_AREA                       <chr> "45", "365", "138", "439", "203...
## $ CPD_NEIGHBORHOOD               <chr> "WALNUT HILLS", "AVONDALE", "MA...
## $ SNA_NEIGHBORHOOD               <chr> "WALNUT HILLS", "AVONDALE", "MA...
## $ WEAPONS                        <chr> "99 - NONE", "40--PERSONAL WEAP...
## $ DATE_OF_CLEARANCE              <chr> "03/28/2015 12:00:00 AM +0000",...
## $ HOUR_FROM                      <int> 152, 1020, 240, 1535, 1645, 143...
## $ HOUR_TO                        <int> 155, 1030, 244, 1540, 1712, 144...
## $ ADDRESS_X                      <chr> "21XX FULTON AV", "34XX READING...
## $ LONGITUDE_X                    <dbl> -84.49080, -84.49155, -84.39439...
## $ LATITUDE_X                     <dbl> 39.11996, 39.14167, 39.15156, 3...
## $ VICTIM_AGE                     <chr> "26-30", "41-50", "UNKNOWN", "1...
## $ VICTIM_RACE                    <chr> "BLACK", "BLACK", "", "BLACK", ...
## $ VICTIM_ETHNICITY               <chr> "NOT OF HISPANIC ORIG", "NOT OF...
## $ VICTIM_GENDER                  <chr> "FEMALE", "MALE", "", "MALE", "...
## $ SUSPECT_AGE                    <chr> "51-60", "UNKNOWN", "18-25", "1...
## $ SUSPECT_RACE                   <chr> "BLACK", "BLACK", "BLACK", "BLA...
## $ SUSPECT_ETHNICITY              <chr> "NOT OF HISPANIC ORIG", "NOT OF...
## $ SUSPECT_GENDER                 <chr> "MALE", "MALE", "FEMALE", "MALE...
## $ TOTALNUMBERVICTIMS             <int> 1, 1, 1, 2, 1, 2, 1, 2, 1, 2, 2...
## $ TOTALSUSPECTS                  <int> 1, 1, 2, 4, NA, 2, 1, 2, NA, NA...
## $ UCR_GROUP                      <chr> "PART 2 MINOR", "PART 2 MINOR",...
## $ COMMUNITY_COUNCIL_NEIGHBORHOOD <chr> "WALNUT HILLS", "AVONDALE", "MA...
## $ ZIP                            <dbl> 2.233473e-319, 2.234115e-319, 2...

• Examine data - In order to clean the data, it is important to check whether there are any missing values or NA values. The “summary” function will help us to identify whether there are any NA or missing values for any column.

summary(crime_data)

##   INSTANCEID        INCIDENT_NO        DATE_REPORTED     
##  Length:355379      Length:355379      Length:355379     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##   DATE_FROM           DATE_TO              CLSD                UCR        
##  Length:355379      Length:355379      Length:355379      Min.   :   0.0  
##  Class :character   Class :character   Class :character   1st Qu.: 552.0  
##  Mode  :character   Mode  :character   Mode  :character   Median : 600.0  
##                                                           Mean   : 803.9  
##                                                           3rd Qu.: 810.0  
##                                                           Max.   :2761.0  
##                                                           NA's   :71      
##      DST                BEAT             OFFENSE         
##  Length:355379      Length:355379      Length:355379     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##    LOCATION          THEFT_CODE           FLOOR          
##  Length:355379      Length:355379      Length:355379     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##      SIDE             OPENING           HATE_BIAS        
##  Length:355379      Length:355379      Length:355379     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##   DAYOFWEEK           RPT_AREA         CPD_NEIGHBORHOOD  
##  Length:355379      Length:355379      Length:355379     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##  SNA_NEIGHBORHOOD     WEAPONS          DATE_OF_CLEARANCE    HOUR_FROM     
##  Length:355379      Length:355379      Length:355379      Min.   :   0.0  
##  Class :character   Class :character   Class :character   1st Qu.: 130.0  
##  Mode  :character   Mode  :character   Mode  :character   Median : 230.0  
##                                                           Mean   : 816.9  
##                                                           3rd Qu.:1623.0  
##                                                           Max.   :2359.0  
##                                                           NA's   :7       
##     HOUR_TO        ADDRESS_X          LONGITUDE_X       LATITUDE_X   
##  Min.   :   0.0   Length:355379      Min.   :-84.82   Min.   :39.05  
##  1st Qu.: 145.0   Class :character   1st Qu.:-84.57   1st Qu.:39.12  
##  Median : 815.0   Mode  :character   Median :-84.52   Median :39.14  
##  Mean   : 922.3                      Mean   :-84.52   Mean   :39.14  
##  3rd Qu.:1651.0                      3rd Qu.:-84.49   3rd Qu.:39.16  
##  Max.   :2359.0                      Max.   :-84.25   Max.   :39.36  
##  NA's   :1316                        NA's   :45864    NA's   :45864  
##   VICTIM_AGE        VICTIM_RACE        VICTIM_ETHNICITY  
##  Length:355379      Length:355379      Length:355379     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##  VICTIM_GENDER      SUSPECT_AGE        SUSPECT_RACE      
##  Length:355379      Length:355379      Length:355379     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##  SUSPECT_ETHNICITY  SUSPECT_GENDER     TOTALNUMBERVICTIMS TOTALSUSPECTS   
##  Length:355379      Length:355379      Min.   :  1.000    Min.   : 1.00   
##  Class :character   Class :character   1st Qu.:  1.000    1st Qu.: 1.00   
##  Mode  :character   Mode  :character   Median :  1.000    Median : 1.00   
##                                        Mean   :  1.433    Mean   : 1.64   
##                                        3rd Qu.:  1.000    3rd Qu.: 2.00   
##                                        Max.   :127.000    Max.   :16.00   
##                                        NA's   :146        NA's   :194149  
##   UCR_GROUP         COMMUNITY_COUNCIL_NEIGHBORHOOD      ZIP   
##  Length:355379      Length:355379                  Min.   :0  
##  Class :character   Class :character               1st Qu.:0  
##  Mode  :character   Mode  :character               Median :0  
##                                                    Mean   :0  
##                                                    3rd Qu.:0  
##                                                    Max.   :0  
## 

We can see that there are many NA’s in UCR, BEAT, RPT_AREA, LONGITUDE_X, LATITUDE_X, TOTALNUMBERVICTIMS, TOTALSUSPECTS and ZIP.

• Clean data - It is always a best practice to clean the data before starting the analysis otherwise due to missing or NA values in the dataset, we might land up with biased and incorrect analysis.

• Identify the NA values - Below R code provides the total count of NA values in this particular dataset.

sum(is.na(crime_data))

## [1] 287417

• Remove all the NA values - Before proceeding further with any kind of analysis, it will be best to remove all the NA values from the dataset in order ti get a clean, consistent, comprehensive, conform, consistent and current dataset.

crime_data_cleaned <- na.omit(crime_data)

• Check dataset - We will check the dataset once at this point in order to be sure that all the NA values have been removed successfully.

sum(is.na(crime_data_cleaned))

## [1] 0

summary(crime_data_cleaned)

##   INSTANCEID        INCIDENT_NO        DATE_REPORTED     
##  Length:137065      Length:137065      Length:137065     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##   DATE_FROM           DATE_TO              CLSD                UCR        
##  Length:137065      Length:137065      Length:137065      Min.   :   0.0  
##  Class :character   Class :character   Class :character   1st Qu.: 551.0  
##  Mode  :character   Mode  :character   Mode  :character   Median : 701.0  
##                                                           Mean   : 844.2  
##                                                           3rd Qu.: 862.0  
##                                                           Max.   :2761.0  
##      DST                BEAT             OFFENSE         
##  Length:137065      Length:137065      Length:137065     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##    LOCATION          THEFT_CODE           FLOOR          
##  Length:137065      Length:137065      Length:137065     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##      SIDE             OPENING           HATE_BIAS        
##  Length:137065      Length:137065      Length:137065     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##   DAYOFWEEK           RPT_AREA         CPD_NEIGHBORHOOD  
##  Length:137065      Length:137065      Length:137065     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##  SNA_NEIGHBORHOOD     WEAPONS          DATE_OF_CLEARANCE    HOUR_FROM   
##  Length:137065      Length:137065      Length:137065      Min.   :   0  
##  Class :character   Class :character   Class :character   1st Qu.: 140  
##  Mode  :character   Mode  :character   Mode  :character   Median : 645  
##                                                           Mean   : 913  
##                                                           3rd Qu.:1730  
##                                                           Max.   :2359  
##     HOUR_TO      ADDRESS_X          LONGITUDE_X       LATITUDE_X   
##  Min.   :   0   Length:137065      Min.   :-84.82   Min.   :39.05  
##  1st Qu.: 170   Class :character   1st Qu.:-84.57   1st Qu.:39.12  
##  Median :1055   Mode  :character   Median :-84.52   Median :39.14  
##  Mean   :1027                      Mean   :-84.52   Mean   :39.14  
##  3rd Qu.:1820                      3rd Qu.:-84.49   3rd Qu.:39.16  
##  Max.   :2359                      Max.   :-84.26   Max.   :39.36  
##   VICTIM_AGE        VICTIM_RACE        VICTIM_ETHNICITY  
##  Length:137065      Length:137065      Length:137065     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##  VICTIM_GENDER      SUSPECT_AGE        SUSPECT_RACE      
##  Length:137065      Length:137065      Length:137065     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##  SUSPECT_ETHNICITY  SUSPECT_GENDER     TOTALNUMBERVICTIMS TOTALSUSPECTS   
##  Length:137065      Length:137065      Min.   : 1.000     Min.   : 1.000  
##  Class :character   Class :character   1st Qu.: 1.000     1st Qu.: 1.000  
##  Mode  :character   Mode  :character   Median : 1.000     Median : 1.000  
##                                        Mean   : 1.375     Mean   : 1.611  
##                                        3rd Qu.: 1.000     3rd Qu.: 2.000  
##                                        Max.   :15.000     Max.   :16.000  
##   UCR_GROUP         COMMUNITY_COUNCIL_NEIGHBORHOOD      ZIP   
##  Length:137065      Length:137065                  Min.   :0  
##  Class :character   Class :character               1st Qu.:0  
##  Mode  :character   Mode  :character               Median :0  
##                                                    Mean   :0  
##                                                    3rd Qu.:0  
##                                                    Max.   :0

• Prepare dataset - At this stage, we will trim down the dataset to only those columns with which we should work further for our analysis.

crime_data_final <- select(crime_data_cleaned, 10, 28, 29, 31)
summary(crime_data_final)

##    OFFENSE           VICTIM_AGE        VICTIM_RACE       
##  Length:137065      Length:137065      Length:137065     
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##  VICTIM_GENDER     
##  Length:137065     
##  Class :character  
##  Mode  :character

• Factorizing variables - Factor variables are categorical variables that can be either numeric or string variables. There are a number of advantages to converting categorical variables to factor variables. The most important advantage is that they can be used in statistical modeling where they will be implemented correctly, i.e., they will then be assigned the correct number of degrees of freedom. Factor variables are also very useful in many different types of graphics. To create a factor variable we use the factor function. The only required argument is a vector of values which can be either string or numeric.

attach(crime_data_final)
VICTIM_AGE <- as.factor(VICTIM_AGE)
VICTIM_RACE <- as.factor(VICTIM_RACE)
VICTIM_GENDER <- as.factor(VICTIM_GENDER)
OFFENSE <- as.factor(OFFENSE)

• Test:Train dataset - We divided our data into train: test in the ratio 70:30 so that we do not get biased results and can compare the trained data model with the test data model.

smp_size <- floor(0.7 * nrow(crime_data_final))
set.seed(25000)
train <- sample(seq_len(nrow(crime_data_final)), size = smp_size)
crime_train_data <- crime_data_final[train,]
crime_test_data <- crime_data_final[-train,]
summary(crime_train_data)

##    OFFENSE           VICTIM_AGE        VICTIM_RACE       
##  Length:95945       Length:95945       Length:95945      
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##  VICTIM_GENDER     
##  Length:95945      
##  Class :character  
##  Mode  :character

summary(crime_test_data)

##    OFFENSE           VICTIM_AGE        VICTIM_RACE       
##  Length:41120       Length:41120       Length:41120      
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##  VICTIM_GENDER     
##  Length:41120      
##  Class :character  
##  Mode  :character

At this point we are good to go with the dataset since it is absolutely clean.

Proposed Exploratory Data Analysis:

From this above dataset, my target is to analyse the crime based upon gender, age, race and ethinicity. With Decision tree analysis and Random Forest analysis, I feel the outcome will be proper based on our categorical response variable. My target is to get an outcome where we can observe the crime, age, race and ethinicity which is more prevalent amongst men and women in Cincinnati. For example, expecting outcomes like below:

Male - between age 30-40, race - Black, ethinicity - Black American and most prevalent offense is robbery or

Female - between age 18-30, race - American, ethinicity - American, most prevalent crime is sexual harassment.