akhilub/YBADINEN_MANISHKR_AKHILUB_MOTHIPRA_ChicagoCrime

hw7-api-manishkr97 created by GitHub Classroom

Jupyter Notebook

City of Chicago crime data visualization

Authors: Yeshwanth Badineni, Akhil Singh Chauhan, Manish katiki, and Mothi Kopparla

Introduction

The code that is being imported is of 'json'(Java Script Object Notation) format.
Data is taken from Chicago Data Portal.
This dataset reflects reported incidents of crime that occurred in the City of Chicago from 2001 to present, minus the most recent seven days.

Sources

The code retrieves data from City of Chicago | Data Portal

Explanation of the Code

The code, API_Chicago_Crime.ipynb, begins by importing necessary Python packages:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import folium
import seaborn as sns
from folium.plugins import MarkerCluster

NOTE: You may need to install folium.
You may do that by following the code below:

pip install folium

We then import data from the above mentioned source:

url = 'https://data.cityofchicago.org/resource/ijzp-q8t2.json?$limit=100000&'
df = pd.read_json(url)	
df

We then clean our data from some unwanted columns present in the data as follows:

drop = [':@computed_region_awaf_s7ux',':@computed_region_6mkv_f3dw',':@computed_region_vrxf_vc4k',':@computed_region_bdys_3d7i',':@computed_region_43wa_7qmu',':@computed_region_rpca_8um6',':@computed_region_d9mm_jgwp',':@computed_region_d3ds_rm58']
df.drop(drop, inplace=True, axis=1)

we visualize the data, No.of crimes according to it's category:

plt.figure(figsize=(15,6))
df['primary_type'].value_counts().plot.bar()
plt.title("Crimes")
plt.show()

The output from this code is shown below:

we visualize the data, No.of crimes according to it's location:

plt.figure(figsize = (15, 10))
sns.countplot(y= 'location_description', data = df, order = df['location_description'].value_counts().iloc[:10].index);

The output from this code is shown below:

Plotting the communities in Chicago city :

for i in range(len(new_locations)):
    lat = new_locations.iloc[i][0]
    long = new_locations.iloc[i][1]
    popup_text = """Community Index : {}<br>
                Arrest : {}<br>
                Location Description : {}<br>"""
    popup_text = popup_text.format(new_locations.index[i],
                               new_locations.iloc[i][-1],
                               new_locations.iloc[i][-2]
                               )
    folium.CircleMarker(location = [lat, long], popup= popup_text, fill = True).add_to(chicago_map)

The output from this code is shown below:

Visualizing the density of crimes in communities:

for i in range(500):
    lat = CR_index['LocationCoord'].iloc[i][0]
    long = CR_index['LocationCoord'].iloc[i][1]
    radius = CR_index['ValueCount'].iloc[i] /5
    
    if CR_index['ValueCount'].iloc[i] >75:
        color = "#FF4500"
    else:
        color = "#008080"
    
    popup_text = """Latitude : {}<br>Longitude : {}<br>Criminal Incidents : {}<br>"""
    
    popup_text = popup_text.format(lat,long,CR_index['ValueCount'].iloc[i])
                               
    folium.CircleMarker(location = [lat, long], popup= popup_text,radius = radius, color = color, fill = True)\
    .add_to(chicago_map_crime)

The output from this code is shown below:

Assaults commited with guns by location:

mc = MarkerCluster()
for row in gun_Battery_df.itertuples():
    mc.add_child(folium.Marker(location=[row.latitude,  row.longitude], popup= 'gun'))
my_map1.add_child(mc)

The output from this code is shown below:

How to Run the Code

Launch Jupyter Notebook on Anaconda.
Open and run API_Chicago_Crime.ipynb cell by cell.

Note: Uncomment the pip install folium , if the package is not pre-installed on the system.

You can interact with the maps to zoom in and zoom out.

Suggestions

We can further use predictive modeling to predict the crime at a recurring scene and can even suggest more patrolling at specific timings so as to curb crimes at a faster pace. Also studying common types of crimes commited in a location can allow police to take appropriate steps to mitigate them.

References