Reference Code

Index

• Data Collection Using API

• Data Munging

• Visualizations

  • Barplot Drilldown
  • Barplot Trending
  • Calendar Plot
  • Radar Plot
  • Linked Plot
  • Geomap_Community
  • Geomap_Block
  • Treemap

Data Collection Using API

Return to Index

#!pip install sodapy

import pandas as pd
import numpy as np
from sodapy import Socrata
import time
import warnings
warnings.filterwarnings('ignore')

client = Socrata("data.cityofchicago.org",
                 app_token="Your App Token",
                 username="example@domain.com",
                 password="Your Password")

start = time.time()
results = client.get("ijzp-q8t2", limit=5000000, where = "year>2015")
end = time.time()

print(end-start)

results_df = pd.DataFrame.from_records(results)

set(results_df["year"])

results_df.to_csv("Chicago_Crime_16_21.csv", index = False)

Data Munging

Return to Index

data = pd.read_csv("Chicago_Crime_16_21.csv")
data.head(3)

	id	case_number	date	block	iucr	primary_type	description	arrest	domestic	beat	...	community_area	fbi_code	year	updated_on	location_description	x_coordinate	y_coordinate	latitude	longitude	location
0	11645836	JC212333	2016-05-01T00:25:00.000	055XX S ROCKWELL ST	1153	DECEPTIVE PRACTICE	FINANCIAL IDENTITY THEFT OVER $ 300	False	False	824	...	63.0	11	2016	2019-04-06T16:04:43.000	NaN	NaN	NaN	NaN	NaN	NaN
1	11043021	JA367631	2016-10-19T19:00:00.000	075XX S YATES BLVD	0610	BURGLARY	FORCIBLE ENTRY	False	False	421	...	43.0	05	2016	2017-08-05T15:50:08.000	RESTAURANT	NaN	NaN	NaN	NaN	NaN
2	11243066	JB168427	2016-03-29T07:00:00.000	067XX S RIDGELAND AVE	1153	DECEPTIVE PRACTICE	FINANCIAL IDENTITY THEFT OVER $ 300	False	False	332	...	43.0	11	2016	2018-03-01T15:54:55.000	OTHER	NaN	NaN	NaN	NaN	NaN

3 rows × 22 columns

data.date = data.date.str.replace("T", " ")
data.date = pd.to_datetime(data.date, format="%Y-%m-%d %H:%M:%S.%f")
data.updated_on = data.updated_on.str.replace("T", " ")
data.updated_on = pd.to_datetime(data.updated_on, format="%Y-%m-%d %H:%M:%S.%f")

data.dropna(how="any", inplace=True)
data["hour"] = data.date.dt.hour
data=data[data.year!=2022]

data.district = data.district.astype("int")

data.community_area=data.community_area.astype("int")

data.drop(columns=["id","case_number", "iucr", "beat", "ward", "fbi_code", "x_coordinate", "y_coordinate","location"], inplace=True)

data.reset_index(drop=True, inplace=True)

data.shape

(1461450, 14)

data.primary_type.replace("CRIM SEXUAL ASSAULT", "CRIMINAL SEXUAL ASSAULT", inplace=True)
data.primary_type.replace("NON-CRIMINAL (SUBJECT SPECIFIED)", "NON-CRIMINAL", inplace=True)
data.primary_type.replace("NON - CRIMINAL", "NON-CRIMINAL", inplace=True)
data.primary_type.replace("CRIMINAL SEXUAL ASSAULT", "SEX OFFENSE", inplace=True)

data.to_csv("crime_clean.csv", index=False)

Visualizations

Return to Index

import matplotlib.pyplot as plt
import seaborn as sns
import plotly.graph_objects as go
import plotly.express as px

data=pd.read_csv('crime_clean.csv')
data.head()

	date	block	primary_type	description	arrest	domestic	district	community_area	year	updated_on	location_description	latitude	longitude	hour
0	2016-05-03 09:45:00	055XX W VAN BUREN ST	ASSAULT	AGGRAVATED - HANDGUN	False	False	15	25	2016	2020-03-25 15:45:43	STREET	41.874758	-87.762379	9
1	2016-05-19 00:00:00	026XX W NORTH AVE	BATTERY	AGGRAVATED - HANDGUN	False	False	14	24	2016	2020-03-25 15:45:43	STREET	41.910261	-87.692759	0
2	2016-05-22 21:25:00	057XX W BLOOMINGDALE AVE	BATTERY	AGGRAVATED - HANDGUN	False	False	25	25	2016	2020-03-25 15:45:43	PARK PROPERTY	41.912628	-87.768864	21
3	2016-06-01 01:22:00	013XX W LELAND AVE	BATTERY	AGGRAVATED - HANDGUN	False	False	19	3	2016	2020-03-25 15:45:43	SIDEWALK	41.967145	-87.663292	1
4	2016-06-10 15:00:00	007XX S HOMAN AVE	ASSAULT	AGGRAVATED - HANDGUN	False	False	11	27	2016	2020-03-25 15:45:43	STREET	41.872253	-87.710733	15

Barplot Drilldown

Return to Index

#Group by year and count crime types
Count_by_year=pd.DataFrame(data.groupby('year').primary_type.value_counts())

#Reformat the dataframe
Count_by_year['Year']=[year for year,crimetype in Count_by_year.index]
Count_by_year['Crime_Type']=[crimetype for year,crimetype in Count_by_year.index]
Count_by_year.index=range(Count_by_year.shape[0])
Count_by_year.rename(columns={'primary_type':'Count'},inplace=True)
Count_by_year=Count_by_year.loc[:,['Crime_Type','Year','Count']]
Count_by_year.Year=Count_by_year.Year.astype('int')
Crimetypes=Count_by_year.Crime_Type.unique()
Count_by_year

	Crime_Type	Year	Count
0	THEFT	2016	61037
1	BATTERY	2016	50246
2	CRIMINAL DAMAGE	2016	30931
3	ASSAULT	2016	18720
4	DECEPTIVE PRACTICE	2016	17374
...	...	...	...
176	GAMBLING	2021	13
177	HUMAN TRAFFICKING	2021	12
178	NON-CRIMINAL	2021	4
179	PUBLIC INDECENCY	2021	4
180	OTHER NARCOTIC VIOLATION	2021	2

181 rows × 3 columns

Count_by_year.sort_values(by = ["Year", "Count"], ascending=False, inplace=True)

output_list = []
for year in [2016,2017,2018,2019,2020,2021]:
    temp = []
    for i in Count_by_year[Count_by_year.Year==2020].values:
        temp.append([i[0],i[2]])
    output_list.append(temp)
output_list

[[['BATTERY', 41413],
  ['THEFT', 40225],
  ['CRIMINAL DAMAGE', 24694],
  ['ASSAULT', 18207],
  ['DECEPTIVE PRACTICE', 15906],
  ['OTHER OFFENSE', 12396],
  ['MOTOR VEHICLE THEFT', 9896],
  ['BURGLARY', 8703],
  ['WEAPONS VIOLATION', 8416],
  ['ROBBERY', 7847],
  ['NARCOTICS', 7269],
  ['CRIMINAL TRESPASS', 4151],
  ['SEX OFFENSE', 2033],
  ['OFFENSE INVOLVING CHILDREN', 1842],
  ['PUBLIC PEACE VIOLATION', 1266],
  ['HOMICIDE', 791],
  ['INTERFERENCE WITH PUBLIC OFFICER', 654],
  ['ARSON', 587],
  ['PROSTITUTION', 275],
  ['STALKING', 194],
  ['INTIMIDATION', 161],
  ['CONCEALED CARRY LICENSE VIOLATION', 148],
  ['LIQUOR LAW VIOLATION', 140],
  ['KIDNAPPING', 118],
  ['OBSCENITY', 50],
  ['GAMBLING', 25],
  ['PUBLIC INDECENCY', 9],
  ['OTHER NARCOTIC VIOLATION', 6],
  ['HUMAN TRAFFICKING', 5],
  ['NON-CRIMINAL', 1],
  ['RITUALISM', 1]],
 [['BATTERY', 41413],
  ['THEFT', 40225],
  ['CRIMINAL DAMAGE', 24694],
  ['ASSAULT', 18207],
  ['DECEPTIVE PRACTICE', 15906],
  ['OTHER OFFENSE', 12396],
  ['MOTOR VEHICLE THEFT', 9896],
  ['BURGLARY', 8703],
  ['WEAPONS VIOLATION', 8416],
  ['ROBBERY', 7847],
  ['NARCOTICS', 7269],
  ['CRIMINAL TRESPASS', 4151],
  ['SEX OFFENSE', 2033],
  ['OFFENSE INVOLVING CHILDREN', 1842],
  ['PUBLIC PEACE VIOLATION', 1266],
  ['HOMICIDE', 791],
  ['INTERFERENCE WITH PUBLIC OFFICER', 654],
  ['ARSON', 587],
  ['PROSTITUTION', 275],
  ['STALKING', 194],
  ['INTIMIDATION', 161],
  ['CONCEALED CARRY LICENSE VIOLATION', 148],
  ['LIQUOR LAW VIOLATION', 140],
  ['KIDNAPPING', 118],
  ['OBSCENITY', 50],
  ['GAMBLING', 25],
  ['PUBLIC INDECENCY', 9],
  ['OTHER NARCOTIC VIOLATION', 6],
  ['HUMAN TRAFFICKING', 5],
  ['NON-CRIMINAL', 1],
  ['RITUALISM', 1]],
 [['BATTERY', 41413],
  ['THEFT', 40225],
  ['CRIMINAL DAMAGE', 24694],
  ['ASSAULT', 18207],
  ['DECEPTIVE PRACTICE', 15906],
  ['OTHER OFFENSE', 12396],
  ['MOTOR VEHICLE THEFT', 9896],
  ['BURGLARY', 8703],
  ['WEAPONS VIOLATION', 8416],
  ['ROBBERY', 7847],
  ['NARCOTICS', 7269],
  ['CRIMINAL TRESPASS', 4151],
  ['SEX OFFENSE', 2033],
  ['OFFENSE INVOLVING CHILDREN', 1842],
  ['PUBLIC PEACE VIOLATION', 1266],
  ['HOMICIDE', 791],
  ['INTERFERENCE WITH PUBLIC OFFICER', 654],
  ['ARSON', 587],
  ['PROSTITUTION', 275],
  ['STALKING', 194],
  ['INTIMIDATION', 161],
  ['CONCEALED CARRY LICENSE VIOLATION', 148],
  ['LIQUOR LAW VIOLATION', 140],
  ['KIDNAPPING', 118],
  ['OBSCENITY', 50],
  ['GAMBLING', 25],
  ['PUBLIC INDECENCY', 9],
  ['OTHER NARCOTIC VIOLATION', 6],
  ['HUMAN TRAFFICKING', 5],
  ['NON-CRIMINAL', 1],
  ['RITUALISM', 1]],
 [['BATTERY', 41413],
  ['THEFT', 40225],
  ['CRIMINAL DAMAGE', 24694],
  ['ASSAULT', 18207],
  ['DECEPTIVE PRACTICE', 15906],
  ['OTHER OFFENSE', 12396],
  ['MOTOR VEHICLE THEFT', 9896],
  ['BURGLARY', 8703],
  ['WEAPONS VIOLATION', 8416],
  ['ROBBERY', 7847],
  ['NARCOTICS', 7269],
  ['CRIMINAL TRESPASS', 4151],
  ['SEX OFFENSE', 2033],
  ['OFFENSE INVOLVING CHILDREN', 1842],
  ['PUBLIC PEACE VIOLATION', 1266],
  ['HOMICIDE', 791],
  ['INTERFERENCE WITH PUBLIC OFFICER', 654],
  ['ARSON', 587],
  ['PROSTITUTION', 275],
  ['STALKING', 194],
  ['INTIMIDATION', 161],
  ['CONCEALED CARRY LICENSE VIOLATION', 148],
  ['LIQUOR LAW VIOLATION', 140],
  ['KIDNAPPING', 118],
  ['OBSCENITY', 50],
  ['GAMBLING', 25],
  ['PUBLIC INDECENCY', 9],
  ['OTHER NARCOTIC VIOLATION', 6],
  ['HUMAN TRAFFICKING', 5],
  ['NON-CRIMINAL', 1],
  ['RITUALISM', 1]],
 [['BATTERY', 41413],
  ['THEFT', 40225],
  ['CRIMINAL DAMAGE', 24694],
  ['ASSAULT', 18207],
  ['DECEPTIVE PRACTICE', 15906],
  ['OTHER OFFENSE', 12396],
  ['MOTOR VEHICLE THEFT', 9896],
  ['BURGLARY', 8703],
  ['WEAPONS VIOLATION', 8416],
  ['ROBBERY', 7847],
  ['NARCOTICS', 7269],
  ['CRIMINAL TRESPASS', 4151],
  ['SEX OFFENSE', 2033],
  ['OFFENSE INVOLVING CHILDREN', 1842],
  ['PUBLIC PEACE VIOLATION', 1266],
  ['HOMICIDE', 791],
  ['INTERFERENCE WITH PUBLIC OFFICER', 654],
  ['ARSON', 587],
  ['PROSTITUTION', 275],
  ['STALKING', 194],
  ['INTIMIDATION', 161],
  ['CONCEALED CARRY LICENSE VIOLATION', 148],
  ['LIQUOR LAW VIOLATION', 140],
  ['KIDNAPPING', 118],
  ['OBSCENITY', 50],
  ['GAMBLING', 25],
  ['PUBLIC INDECENCY', 9],
  ['OTHER NARCOTIC VIOLATION', 6],
  ['HUMAN TRAFFICKING', 5],
  ['NON-CRIMINAL', 1],
  ['RITUALISM', 1]],
 [['BATTERY', 41413],
  ['THEFT', 40225],
  ['CRIMINAL DAMAGE', 24694],
  ['ASSAULT', 18207],
  ['DECEPTIVE PRACTICE', 15906],
  ['OTHER OFFENSE', 12396],
  ['MOTOR VEHICLE THEFT', 9896],
  ['BURGLARY', 8703],
  ['WEAPONS VIOLATION', 8416],
  ['ROBBERY', 7847],
  ['NARCOTICS', 7269],
  ['CRIMINAL TRESPASS', 4151],
  ['SEX OFFENSE', 2033],
  ['OFFENSE INVOLVING CHILDREN', 1842],
  ['PUBLIC PEACE VIOLATION', 1266],
  ['HOMICIDE', 791],
  ['INTERFERENCE WITH PUBLIC OFFICER', 654],
  ['ARSON', 587],
  ['PROSTITUTION', 275],
  ['STALKING', 194],
  ['INTIMIDATION', 161],
  ['CONCEALED CARRY LICENSE VIOLATION', 148],
  ['LIQUOR LAW VIOLATION', 140],
  ['KIDNAPPING', 118],
  ['OBSCENITY', 50],
  ['GAMBLING', 25],
  ['PUBLIC INDECENCY', 9],
  ['OTHER NARCOTIC VIOLATION', 6],
  ['HUMAN TRAFFICKING', 5],
  ['NON-CRIMINAL', 1],
  ['RITUALISM', 1]]]

Barplot Trending

Return to Index

Count_by_year.sort_values(by = ["Year","Count"], ascending=[True,False], inplace=True)

Count_by_year

	Crime_Type	Year	Count
0	THEFT	2016	61037
1	BATTERY	2016	50246
2	CRIMINAL DAMAGE	2016	30931
3	ASSAULT	2016	18720
4	DECEPTIVE PRACTICE	2016	17374
...	...	...	...
176	GAMBLING	2021	13
177	HUMAN TRAFFICKING	2021	12
178	NON-CRIMINAL	2021	4
179	PUBLIC INDECENCY	2021	4
180	OTHER NARCOTIC VIOLATION	2021	2

181 rows × 3 columns

fig=px.bar(
    Count_by_year,x='Crime_Type',y='Count',animation_frame='Year',
    category_orders={'Crime_Type':Count_by_year.Crime_Type.unique()},
    color='Count',color_continuous_scale='Agsunset_r',log_y=True)
fig.update_layout(
    font={"family":"sans-serif", "color":"black"},
    title={
        'text': "Crimes Cases of Different Crime Types: 2016-2021",
        'font': dict(size=20),
        'y':0.975,
        'x':0,
        'yanchor': 'top'},
    template="plotly_white",
    yaxis=dict(type="log",title='Crime Cases'),
    xaxis = dict(title='Select Year with Slider', tickangle=-45)
)

fig["layout"].pop("updatemenus")
fig.update(layout_coloraxis_showscale=False)
fig['layout']['sliders'][0]['pad']=dict(r= 10, t= 200,l=-70)

fig.write_html('barplot_trend.html',config={'responsive':False})
fig.show()

Calendar Plot

Return to Index

#!pip install plotly_calplot
#!pip install calmap

from plotly_calplot import calplot
import calmap

data.date = pd.to_datetime(data.date, format="%Y-%m-%d %H:%M:%S.%f")
data["day"] = data["date"].dt.date

df_calendar = data[data.primary_type.isin(["THEFT","BATTERY","CRIMINAL DAMAGE","ASSAULT", "ROBBERY"])][["day","date"]]\
.groupby(["day"]).count().reset_index().rename(columns={"date":"count"})

df_calendar.day = pd.to_datetime(df_calendar.day, format="%Y-%m-%d")
df_calendar

	day	count
0	2016-01-01	533
1	2016-01-02	360
2	2016-01-03	435
3	2016-01-04	383
4	2016-01-05	405
...	...	...
2187	2021-12-27	319
2188	2021-12-28	292
2189	2021-12-29	304
2190	2021-12-30	350
2191	2021-12-31	351

2192 rows × 2 columns

fig = go.Figure(data=calplot(
         df_calendar,
         x="day",
         y="count",
         years_title=True,
         title = 'Crime Case Calendar of Chicago: 2016-2021 ',
         dark_theme=False,
         month_lines_width=3, 
         month_lines_color="#fff",
         colorscale= "Agsunset_r"
        
))
fig.update_traces(selector=dict(type="heatmap"), zmax=df_calendar["count"].max(), zmin=df_calendar["count"].min())

fig.update_traces(
        showscale=True,
        selector=dict(type="heatmap"),
)

fig.update_layout(
    title=dict(font=dict(size=20,
                        family="sans-serif",
                        color="black")),
    title_x=0,
    title_y=0.98,
    margin = dict(t=70,b=50),
    hoverlabel = dict(
        bgcolor = "white",
        font_size = 13,
        font_family = "sans-serif"
    )
)

fig.show()
fig.write_html("calendar_plot.html",config={'responsive':False})

Radar Plot

Return to Index

df_radar = data[["hour","date","primary_type"]]\
.groupby(["hour","primary_type"]).count().reset_index().rename(columns={"date":"count"})

df_radar.hour = pd.DataFrame(["0"+str(i) if len(str(i))==1 else str(i) for i in df_radar.hour])
df_radar

	hour	primary_type	count
0	00	ARSON	205
1	00	ASSAULT	3859
2	00	BATTERY	14428
3	00	BURGLARY	3065
4	00	CONCEALED CARRY LICENSE VIOLATION	42
...	...	...	...
693	23	ROBBERY	3227
694	23	SEX OFFENSE	716
695	23	STALKING	48
696	23	THEFT	10420
697	23	WEAPONS VIOLATION	2941

698 rows × 3 columns

palette=["#9b5de5","#c65ccd","#f15bb5","#f8a07b","#fbc25e","#fee440","#7fd09d","#00bbf9","#00d8e7","#00f5d4"]

layout=go.Layout(polar=dict(angularaxis = dict(rotation = 90,direction = "clockwise",tickfont=dict(size=15))))
fig = go.Figure(data=[
    go.Scatterpolar(
        name = "THEFT",
        r=df_radar[df_radar.primary_type=="THEFT"]["count"].append(df_radar[(df_radar.primary_type=="THEFT")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="THEFT"]["hour"].append(df_radar[(df_radar.primary_type=="THEFT")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[0], shape="spline", smoothing=1)),
    go.Scatterpolar(
        name = "BATTERY",
        r=df_radar[df_radar.primary_type=="BATTERY"]["count"].append(df_radar[(df_radar.primary_type=="BATTERY")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="BATTERY"]["hour"].append(df_radar[(df_radar.primary_type=="BATTERY")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[1], shape="spline", smoothing=0.7)),
    go.Scatterpolar(
        name = "CRIMINAL DAMAGE",
        r=df_radar[df_radar.primary_type=="CRIMINAL DAMAGE"]["count"].append(df_radar[(df_radar.primary_type=="CRIMINAL DAMAGE")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="CRIMINAL DAMAGE"]["hour"].append(df_radar[(df_radar.primary_type=="CRIMINAL DAMAGE")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[2], shape="spline", smoothing=0.7)),
    go.Scatterpolar(
        name = "ASSAULT",
        r=df_radar[df_radar.primary_type=="ASSAULT"]["count"].append(df_radar[(df_radar.primary_type=="ASSAULT")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="ASSAULT"]["hour"].append(df_radar[(df_radar.primary_type=="ASSAULT")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[3], shape="spline", smoothing=0.7)),
    go.Scatterpolar(
        name = "ROBBERY",
        r=df_radar[df_radar.primary_type=="ROBBERY"]["count"].append(df_radar[(df_radar.primary_type=="ROBBERY")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="ROBBERY"]["hour"].append(df_radar[(df_radar.primary_type=="ROBBERY")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[4], shape="spline", smoothing=0.7)),
    go.Scatterpolar(
        name = "DECEPTIVE PRACTICE",
        r=df_radar[df_radar.primary_type=="DECEPTIVE PRACTICE"]["count"].append(df_radar[(df_radar.primary_type=="DECEPTIVE PRACTICE")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="DECEPTIVE PRACTICE"]["hour"].append(df_radar[(df_radar.primary_type=="DECEPTIVE PRACTICE")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[5], shape="spline", smoothing=0.7)), 
    go.Scatterpolar(
        name = "NARCOTICS",
        r=df_radar[df_radar.primary_type=="NARCOTICS"]["count"].append(df_radar[(df_radar.primary_type=="NARCOTICS")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="NARCOTICS"]["hour"].append(df_radar[(df_radar.primary_type=="NARCOTICS")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[6], shape="spline", smoothing=0.7)),
    go.Scatterpolar(
        name = "MOTOR VEHICLE THEFT",
        r=df_radar[df_radar.primary_type=="MOTOR VEHICLE THEFT"]["count"].append(df_radar[(df_radar.primary_type=="MOTOR VEHICLE THEFT")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="MOTOR VEHICLE THEFT"]["hour"].append(df_radar[(df_radar.primary_type=="MOTOR VEHICLE THEFT")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[7], shape="spline", smoothing=0.7)),
    go.Scatterpolar(
        name = "BURGLARY",
        r=df_radar[df_radar.primary_type=="BURGLARY"]["count"].append(df_radar[(df_radar.primary_type=="BURGLARY")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="BURGLARY"]["hour"].append(df_radar[(df_radar.primary_type=="BURGLARY")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[8], shape="spline", smoothing=0.7)),
    go.Scatterpolar(
        name = "WEAPONS VIOLATION",
        r=df_radar[df_radar.primary_type=="WEAPONS VIOLATION"]["count"].append(df_radar[(df_radar.primary_type=="WEAPONS VIOLATION")&(df_radar.hour=="00")]["count"]),
        theta=df_radar[df_radar.primary_type=="WEAPONS VIOLATION"]["hour"].append(df_radar[(df_radar.primary_type=="WEAPONS VIOLATION")&(df_radar.hour=="00")]["hour"]),
        line=dict(color=palette[9], shape="spline", smoothing=0.7))],
                layout=layout)


fig.update_layout(
    updatemenus=[
        dict(font=dict(size=13),
            active=0,
            buttons=list([
                dict(label="ALL",
                     method="update",
                     args=[dict(visible=[True, True, True, True, True, True, True, True, True, True]),
                           dict(title="Top 10 Crime Hourly Distribution: 2016-2021")]),
                dict(label="THEFT",
                     method="update",
                     args=[dict(visible=[True, False, False, False, False, False, False, False, False, False]),
                           dict(title="THEFT Hourly Distribution: 2016-2021")]),
                dict(label="BATTERY",
                     method="update",
                     args=[dict(visible=[False, True, False, False, False, False, False, False, False, False]),
                           dict(title="BATTERY Hourly Distribution: 2016-2021")]),
                dict(label="CRIMINAL DAMAGE",
                     method="update",
                     args=[dict(visible=[False, False, True, False, False, False, False, False, False, False]),
                           dict(title="CRIMINAL DAMAGE Hourly Distribution: 2016-2021")]),
                dict(label="ASSAULT",
                     method="update",
                     args=[dict(visible=[False, False, False, True, False, False, False, False, False, False]),
                           dict(title="ASSAULT Hourly Distribution: 2016-2021")]),
                dict(label="ROBBERY",
                     method="update",
                     args=[dict(visible=[False, False, False, False, True, False, False, False, False, False]),
                           dict(title="ROBBERY Hourly Distribution: 2016-2021")]),
                dict(label="DECEPTIVE PRACTICE",
                     method="update",
                     args=[dict(visible=[False, False, False, False, False, True, False, False, False, False]),
                           dict(title="DECEPTIVE PRACTICE Hourly Distribution: 2016-2021")]),
                dict(label="NARCOTICS",
                     method="update",
                     args=[dict(visible=[False, False, False, False, False, False, True, False, False, False]),
                           dict(title="NARCOTICS Hourly Distribution: 2016-2021")]),
                dict(label="MOTOR VEHICLE THEFT",
                     method="update",
                     args=[dict(visible=[False, False, False, False, False, False, False, True, False, False]),
                           dict(title="MOTOR VEHICLE THEFT Hourly Distribution: 2016-2021")]),
                dict(label="BURGLARY",
                     method="update",
                     args=[dict(visible=[False, False, False, False, False, False, False, False, True, False]),
                           dict(title="BURGLARY Hourly Distribution: 2016-2021")]),
                dict(label="WEAPONS VIOLATION",
                     method="update",
                     args=[dict(visible=[False, False, False, False, False, False, False, False, False, True]),
                           dict(title="WEAPONS VIOLATION Hourly Distribution: 2016-2021")]),
            ]),
        )
    ])

fig.update_layout(
    font={'family':'sans-serif','color':'black'},
    polar=dict(radialaxis=dict(visible=True)),
    showlegend=True,
    template="plotly_white",
    title_text = "Top 10 Crime Hourly Distribution: 2016-2021", 
    title_x=0, 
    legend=dict(font=dict(size=13), yanchor="top",y=0.9,xanchor="left", x=-0.1),
    title=dict(font=dict(size=20)),
    updatemenus=[dict(x=-0.1,xanchor="left")],
    hoverlabel = dict(
        bgcolor = "white",
        font_size = 13,
        font_family = "sans-serif"
    )
)

fig.update_traces(
    mode="lines",
    hovertemplate='Hour: %{theta}<br>Count: %{r}'
)

fig.show()

fig.write_html("radar_plot.html",config={'responsive':False})

Linked Plot

Return to Index

import altair as alt

data_sum = data[["day", "primary_type", "year"]]\
.groupby(["day", "primary_type"]).count()\
.reset_index().rename(columns={"year":"num", "day":"date"})

data_sum["date"] = pd.to_datetime(data_sum.date)

data_sum = data_sum[data_sum.primary_type.isin(["THEFT","BATTERY","CRIMINAL DAMAGE","ASSAULT","ROBBERY"])]
data_sum["day"] = data_sum.date.dt.day
data_sum["month"] = data_sum.date.dt.month

df = data_sum.groupby(["primary_type","month","day"]).mean().reset_index()
df["date"] = "2021-" + df["month"].astype("str") + "-" + df["day"].astype("str")
df = df[df.date!="2021-2-29"]
df.date = pd.to_datetime(df.date)
df

	primary_type	month	day	num	date
0	ASSAULT	1	1	47.333333	2021-01-01
1	ASSAULT	1	2	35.333333	2021-01-02
2	ASSAULT	1	3	45.666667	2021-01-03
3	ASSAULT	1	4	40.166667	2021-01-04
4	ASSAULT	1	5	43.000000	2021-01-05
...	...	...	...	...	...
1825	THEFT	12	27	128.166667	2021-12-27
1826	THEFT	12	28	135.666667	2021-12-28
1827	THEFT	12	29	121.833333	2021-12-29
1828	THEFT	12	30	130.333333	2021-12-30
1829	THEFT	12	31	109.833333	2021-12-31

1825 rows × 5 columns

scale = alt.Scale(domain=["THEFT","BATTERY","CRIMINAL DAMAGE","ASSAULT","ROBBERY"],
                  range=['#9467bd', '#1f77b4', '#aec7e8', '#95d5b2', '#e7ba52'])
color = alt.Color('primary_type:N', scale=scale,legend=alt.Legend(title="Crime Primary Type"))

# We create two selections:
# - a brush that is active on the top panel
# - a multi-click that is active on the bottom panel
brush = alt.selection_interval(encodings=['x'])
#click = alt.selection_multi(encodings=['color'])
click = alt.selection_single(empty='all', fields=['primary_type'])
#click1 = alt.selection_single(fields=['primary_type'], bind='legend')
click2 = alt.selection_multi(encodings=['color'])

base = alt.Chart(df).properties(
    width=550
).add_selection(click)

points = base.mark_line().encode(
    alt.X('monthdate(date):T', title=''),
    alt.Y('num:Q',
        title='Number of Crimes',
        scale=alt.Scale(domain=[0, 250])
    ),
    color=alt.condition(click, color, alt.value('lightgray'))
).properties(
    height=300
)
bars = base.mark_bar().encode(
    x=alt.X('sum(num):Q', 
            title='Yearly Total Crimes'),
    y=alt.Y('primary_type:N', 
            title="",
            sort=["THEFT","BATTERY","CRIMINAL DAMAGE","ASSAULT","ROBBERY"]),
    color=alt.condition(click, color, alt.value('lightgray'))
).transform_filter(
    click
).properties(
    width=260
)

boxs = base.mark_boxplot().encode(
    x=alt.X('num:Q', 
            title='Daily Crimes'),
    y=alt.Y('primary_type:N', 
            #title="Crime Primary Type",
            sort=["THEFT","BATTERY","CRIMINAL DAMAGE","ASSAULT","ROBBERY"],axis=None),
    color=alt.condition(click, color, alt.value('lightgray'))
).transform_filter(
    click
).properties(
    width=260
)

fig = points & (bars | boxs)
fig = fig.properties(title = "Daily Crime Trend - Average of 6 Years: 2016-2021"
).configure_axis(labelColor="black", labelFontSize=13, labelFont="sans-serif", labelFontWeight="normal", 
                 titleColor="black", titleFontSize=15, titleFont="sans-serif", titleFontWeight="normal", grid=False
).configure_legend(labelColor="black", labelFontSize=13, labelFont="sans-serif", labelFontWeight="normal", 
                   titleColor="black", titleFontSize=15, titleFont="sans-serif", titleFontWeight="normal"
).configure_title(color="black", fontSize=20, font="sans-serif", fontWeight="normal"
).configure_view(strokeWidth=0
)

fig.save('linked_plot.html')
fig

Geomap_Community

Return to Index

#!pip install folium
#! pip install geopandas

import folium
import geopandas as gpd
import json

df_agg = data.copy()

community_crime = df_agg.groupby(["year","community_area"]).count().reset_index()[["year","community_area","date"]].\
rename(columns={'community_area':'Community Area', 'date':'Crime Count'})

community_crime['Community Area'] = community_crime['Community Area'].astype('str')

community_crime

	year	Community Area	Crime Count
0	2016	1	3637
1	2016	2	3256
2	2016	3	3515
3	2016	4	1957
4	2016	5	1369
...	...	...	...
457	2021	73	2498
458	2021	74	490
459	2021	75	1713
460	2021	76	1394
461	2021	77	2365

462 rows × 3 columns

com_name = pd.read_csv('CommAreas.csv')
com_name = com_name[['AREA_NUMBE', 'COMMUNITY']]
com_name.columns = ['Community Area', 'Community']
com_name['Community Area'] = com_name['Community Area'].astype('str')

community_crime = pd.merge(community_crime, com_name, on = ['Community Area'], how = 'left')

fh = open('chicago_geo.geojson')
commu_geo_json = json.load(fh)
fh.close

<function TextIOWrapper.close()>

for feature in commu_geo_json['features']:
    feature.update({'id': feature['properties']['area_num_1']})

fig = px.choropleth_mapbox(community_crime, geojson = commu_geo_json, locations = 'Community Area',
                           color = 'Crime Count', animation_frame = 'year',
                           color_continuous_scale = "Agsunset_r",
                           mapbox_style = "carto-positron", range_color = [1000, 8000],
                           zoom = 9, center = {"lat": 41.85, "lon": -87.7},
                           opacity = 0.6, 
                           hover_data = {'Community Area': False,
                                         'Community': True,
                                         'Crime Count': True},
                           title = 'Yearly Crime Cases Map - Communities in Chicago: 2016-2021'
                          )
fig.update_layout(margin={"r":0,"t":80,"l":10,"b":10})
fig.update_layout(
    font_family="sans-serif",
    title = dict(font=dict(size=20, color="black"), x=0.01)
)
fig.update_layout(
    hoverlabel = dict(
        bgcolor = "white",
        font_size = 13,
        font_family = "sans-serif"
    )
)

fig.show()
fig.write_html("geomap_community.html",config={'responsive':False})

Geomap_Block

Return to Index

data['longitude'] = round(data['longitude'],3)
data['latitude'] = round(data['latitude'], 3)

scatter_df = pd.DataFrame(data.groupby(['year', 'latitude', 'longitude']).apply(lambda data: len(data)).reset_index())
scatter_df.columns = ['Year', 'Latitude', 'Longitude', 'Crime Count']
scatter_df

	Year	Latitude	Longitude	Crime Count
0	2016	41.645	-87.615	2
1	2016	41.645	-87.613	1
2	2016	41.645	-87.611	2
3	2016	41.645	-87.602	1
4	2016	41.645	-87.574	1
...	...	...	...	...
199775	2021	42.023	-87.674	16
199776	2021	42.023	-87.673	5
199777	2021	42.023	-87.672	34
199778	2021	42.023	-87.671	6
199779	2021	42.023	-87.666	1

199780 rows × 4 columns

fig = px.density_mapbox(scatter_df, lat = 'Latitude', lon = 'Longitude', animation_frame = 'Year',
                        z = 'Crime Count', radius = 8, center = {"lat": 41.89, "lon": -87.7},
                        zoom = 10, mapbox_style = "carto-positron",
                       color_continuous_scale = "Agsunset_r",
                       range_color = [0, 70],
                       title = 'Yearly Crime Cases Map - Blocks in Chicago: 2016-2021')
fig.update_layout(margin={"r":0,"t":80,"l":10,"b":10})
fig.update_layout(
    font_family="sans-serif",
    title = dict(font=dict(size=20,color="black"), x=0.01)
)
fig.update_layout(
    hoverlabel = dict(
        bgcolor = "white",
        font_size = 13,
        font_family = "sans-serif"
    )
)
fig.show()
fig.write_html("geomap_block.html",config={'responsive':False})

Treemap

Return to Index

Count_by_type=data.groupby(['primary_type','domestic','arrest']).count().reset_index()
Count_by_type=Count_by_type.iloc[:,:4]
Count_by_type.rename(columns={'primary_type':'Crime_Type','id':'Count'},inplace=True)

Count_by_type.arrest.replace(False,'Unarrested',inplace=True)
Count_by_type.arrest.replace(True,'Arrested',inplace=True)

Count_by_type.domestic.replace(False,'Non-domestic',inplace=True)
Count_by_type.domestic.replace(True,'Domestic',inplace=True)

Count_by_type.rename(columns={"date":"Count"}, inplace=True)

Count_by_type.head()

	Crime_Type	domestic	arrest	Count
0	ARSON	Non-domestic	Unarrested	2440
1	ARSON	Non-domestic	Arrested	211
2	ARSON	Domestic	Unarrested	117
3	ARSON	Domestic	Arrested	45
4	ASSAULT	Non-domestic	Unarrested	72509

Count_by_type_shrink = data.groupby("primary_type").count().reset_index()[["primary_type","date"]].sort_values("date",ascending=False).reset_index(drop=True)

to_combine=Count_by_type_shrink.iloc[10:,0]
to_keep=Count_by_type_shrink.iloc[:10,0]

tmp1=Count_by_type.loc[[Count_by_type.Crime_Type[i] in to_keep.to_list() for i in range(Count_by_type.shape[0])],:]
tmp2=Count_by_type.loc[[Count_by_type.Crime_Type[i] in to_combine.to_list() for i in range(Count_by_type.shape[0])],:]
tmp2=tmp2.groupby(['domestic','arrest']).Count.sum().reset_index()
tmp2['Crime_Type']='OTHER TYPES'
tmp2=tmp2.loc[:,['Crime_Type','domestic','arrest','Count']]

Count_by_type_shrink=pd.concat([tmp1,tmp2],ignore_index=False).reset_index(drop=True)
Count_by_type_shrink.head()

	Crime_Type	domestic	arrest	Count
0	ASSAULT	Non-domestic	Unarrested	72509
1	ASSAULT	Non-domestic	Arrested	13668
2	ASSAULT	Domestic	Unarrested	26701
3	ASSAULT	Domestic	Arrested	4497
4	BATTERY	Non-domestic	Unarrested	113186

fig = px.treemap(Count_by_type_shrink, path=['arrest','Crime_Type','domestic'], values='Count',
                 color_discrete_sequence = ['#862ca2', '#edd9b4'])
fig.update_layout(
    font=dict(family="sans-serif"),
    title=dict(
        text="Crime Cases by Arrest, Crime Type and Domestic: 2016-2021",
        font=dict(size=20,color='black'),
        y=0.99,
        x=0,
        yanchor='top'),
    margin=dict(l=16, r=16, t=16, b=16),
    template="plotly_white",
    hoverlabel = dict(
        bgcolor = "white",
        font_size = 13,
        font_family = "sans-serif"
    )
)
fig.update_traces(
    marker=dict(line=dict(color='white',width=1)),
    hovertemplate='%{id}<br>Cases: %{value}'
)
fig.write_html('treemap_arrest.html',config={'responsive':False})
fig.show()

fig = px.treemap(Count_by_type_shrink,path=['Crime_Type','arrest','domestic'], values='Count',
                 color_discrete_sequence = px.colors.sequential.Agsunset
                )
fig.update_layout(
    font=dict(family="sans-serif", color='white'),
    title=dict(
        text="Crime Cases by Crime Type, Arrest and Domestic: 2016-2021",
        font=dict(size=20,color='black'),
        y=0.99,
        x=0,
        yanchor='top'),
    margin=dict(l=16, r=16, t=16, b=16),
    hoverlabel = dict(
        bgcolor = "white",
        font_size = 13,
        font_family = "sans-serif"
    )
)
fig.update_traces(
    marker=dict(line=dict(color='white',width=1)),
    hovertemplate='%{id}<br>Cases: %{value}'
)

fig.write_html('treemap_type.html',config={'responsive':False})
fig.show()