Question: On which days do people travel or move around in which areas are they less likely to be targeted by criminals?

Inspiration: As I was learning about the final project, I recalled an alart email I received from NUPD on Feb.14th about a robbery that occurred inside the Columbus Parking Garage. The victim is my friend's roomate. She was robbed of a large sum of money by three people, two of whom have been arrested so far. This bad timing made me want to use data analysis to know the crime rate and the most common crime categories in Boston in previous years.

Email I got:

The dataset

import pandas as pd
import numpy as np

df_crime = pd.read_csv('Boston Crime Dataset (updated July 2020).csv')

/var/folders/y4/t0fs65x56m39rqs7vzp6m5cr0000gn/T/ipykernel_79157/3857216148.py:1: DtypeWarning: Columns (0,2,6,12) have mixed types. Specify dtype option on import or set low_memory=False.
  df_crime = pd.read_csv('Boston Crime Dataset (updated July 2020).csv')

df_crime

	INCIDENT_NUMBER	OFFENSE_CODE	OFFENSE_CODE_GROUP	OFFENSE_DESCRIPTION	DISTRICT	REPORTING_AREA	SHOOTING	OCCURRED_ON_DATE	YEAR	MONTH	DAY_OF_WEEK	HOUR	UCR_PART	STREET	Lat	Long	Location
0	TESTTEST2	423	NaN	ASSAULT - AGGRAVATED	External		0	2019-10-16 00:00:00	2019	10	Wednesday	0	NaN	RIVERVIEW DR	NaN	NaN	(0.00000000, 0.00000000)
1	S97333701	3301	NaN	VERBAL DISPUTE	C6	915	0	2020-07-18 14:34:00	2020	7	Saturday	14	NaN	MARY BOYLE WAY	42.330813	-71.051368	(42.33081300, -71.05136800)
2	S47513131	2647	NaN	THREATS TO DO BODILY HARM	E18	530	0	2020-06-24 10:15:00	2020	6	Wednesday	10	NaN	READVILLE ST	42.239491	-71.135954	(42.23949100, -71.13595400)
3	I92102201	3301	NaN	VERBAL DISPUTE	E13	583	0	2019-12-20 03:08:00	2019	12	Friday	3	NaN	DAY ST	42.325122	-71.107779	(42.32512200, -71.10777900)
4	I92097173	3115	NaN	INVESTIGATE PERSON	C11	355	0	2019-10-23 00:00:00	2019	10	Wednesday	0	NaN	GIBSON ST	42.297555	-71.059709	(42.29755500, -71.05970900)
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
501065	20224065	3018	NaN	SICK/INJURED/MEDICAL - POLICE	B2	282	0	2020-03-19 07:30:00	2020	3	Thursday	7	NaN	WASHINGTON ST	42.353272	-71.173738	(42.35327200, -71.17373800)
501066	20202856	2672	NaN	BIOLOGICAL THREATS	B2	282	0	2020-03-19 08:30:00	2020	3	Thursday	8	NaN	WARREN ST	42.328234	-71.083289	(42.32823400, -71.08328900)
501067	20046400	1501	NaN	WEAPON VIOLATION - CARRY/ POSSESSING/ SALE/ TR...	B2	330	0	2020-07-02 01:38:00	2020	7	Thursday	1	NaN	PASADENA RD	42.305760	-71.083771	(42.30576000, -71.08377100)
501068	20038446	1501	NaN	WEAPON VIOLATION - CARRY/ POSSESSING/ SALE/ TR...	B2	300	0	2020-06-03 01:15:00	2020	6	Wednesday	1	NaN	WASHINGTON ST	42.323807	-71.089150	(42.32380700, -71.08915000)
501069	20030892	540	NaN	BURGLARY - COMMERICAL	C11	380	0	2020-05-03 00:00:00	2020	5	Sunday	0	NaN	GALLIVAN BLVD	42.283700	-71.047761	(42.28370000, -71.04776100)

501070 rows × 17 columns

Each Features in Dataset means(show in data dictionary):

These FEATURES are all closely related to the problem I'm trying to solve, is sufficient to make progress on my real-world problem described above. For example, there's day_of_week include in teh dataset, if I want to know on which days there is the most crime, I can just take this column.

# get each column's name
print(df_crime.columns.values)

['INCIDENT_NUMBER' 'OFFENSE_CODE' 'OFFENSE_CODE_GROUP'
 'OFFENSE_DESCRIPTION' 'DISTRICT' 'REPORTING_AREA' 'SHOOTING'
 'OCCURRED_ON_DATE' 'YEAR' 'MONTH' 'DAY_OF_WEEK' 'HOUR' 'UCR_PART'
 'STREET' 'Lat' 'Long' 'Location']

# create a dictionary that explains the meaning of each feature present
crime_dict = {'INCIDENT_NUMBER': 'the unique number assigned to a specific person during a specific arrest',
'OFFENSE_CODE': 'the numeric code that corresponds to criminal offenses under Texas State law',
'OFFENSE_CODE_GROUP': 'the offense code group the crime belongs to',
'OFFENSE_DESCRIPTION': 'the description of the offense code group',
'DISTRICT': 'a district where the crime has been committed',
'REPORTING_AREA': 'code for reporting area',
'SHOOTING': 'column depicting if the shooting was involved',
'OCCURRED_ON_DATE': 'date of the crime',
'YEAR': 'the year of crime',
'MONTH': 'the month of crime',
'DAY_OF_WEEK': 'the name of the day of week',
'HOUR': 'the hour of crime(24hr format)',
'UCR_PART': 'FBI ucr part name',
'STREET': 'street name',
'Lat': 'latitude info',
'Long': 'longitude info',
'Location': 'coordinate info'}
crime_dict

{'INCIDENT_NUMBER': 'the unique number assigned to a specific person during a specific arrest',
 'OFFENSE_CODE': 'the numeric code that corresponds to criminal offenses under Texas State law',
 'OFFENSE_CODE_GROUP': 'the offense code group the crime belongs to',
 'OFFENSE_DESCRIPTION': 'the description of the offense code group',
 'DISTRICT': 'a district where the crime has been committed',
 'REPORTING_AREA': 'code for reporting area',
 'SHOOTING': 'column depicting if the shooting was involved',
 'OCCURRED_ON_DATE': 'date of the crime',
 'YEAR': 'the year of crime',
 'MONTH': 'the month of crime',
 'DAY_OF_WEEK': 'the name of the day of week',
 'HOUR': 'the hour of crime(24hr format)',
 'UCR_PART': 'FBI ucr part name',
 'STREET': 'street name',
 'Lat': 'latitude info',
 'Long': 'longitude info',
 'Location': 'coordinate info'}

How the data will be used to solved the problem:

I will find which Boston areas have the highest crime rates and delve into which crime categories are most common in Boston and whether the number of crimes differs between weekdays and weekends. The data will eventually be put together in a graphical presentation through using matpoltlib.