Austin Police Department - Incident Extract

Required packages:

Be sure to have these packages installed in R before running any code! Most of them should be on CRAN and regularly updated, so you should be able to run install.packages("PACKAGE_NAME") to install anything that’s missing!

##   ggplot2     dplyr lubridate  reshape2  magrittr   stringi gridExtra 
##      TRUE      TRUE      TRUE      TRUE      TRUE      TRUE      TRUE 
##     tidyr     ggmap  ggthemes    scales  RSocrata animation 
##      TRUE      TRUE      TRUE      TRUE      TRUE      TRUE

---

Get Data from City of Austin APIs

Using the https://data.austintexas.gov & the Socrata API

APD INCIDENT EXTRACT YTD

APD.file_path   <- "Data/APD_YTD.csv"
APD.socrata_URL <- "https://data.austintexas.gov/resource/b4y9-5x39.csv"

# Import data
APD <- RSocrata::read.socrata(APD.socrata_URL)

# Write as a .csv
write.csv(APD, file = APD.file_path)

Load the dataset for the last 18 months of APD incidents. See description on APD Incidents webpage for more details.

The live-updated Socrata path is https://data.austintexas.gov/resource/b4y9-5x39.csv and the cached comma separated values (CSV) file may be found under the Data/APD_YTD.csv sub-directory.

---

From their website
* Due to the methodological differences in data collection, different data sources may produce different results.
* Our on-line database is continuously being updated. The data provided here represents a particular point in time.
* Updates to the police report database occur daily. Information is available from today’s date back 18 months.
* Due to several factors (once-a-day updates, offense reclassification, reported versus occurred dates, etc.) comparisons should not be made between numbers generated with this database to any other official police reports. * Data provided represents only calls for police service where a report was written.

The Austin Police Department does not assume any liability for any decision made or action taken or not taken by the recipient in reliance upon any information or data provided.

MUNICIPAL COURT CASELOAD INFORMATION

Court.file_path   <- "Data/CourtData.csv"
Court.socrata_URL <- "https://data.austintexas.gov/resource/8jyt-x94k.csv"

# Import data
Court <- RSocrata::read.socrata(Court.socrata_URL)

# Write as a .csv
write.csv(Court, file = Court.file_path)

This data is provided to help with analysis of various violations charged throughout the City of Austin. See Court Cases webpage for more details.

The live-updated Socrata path is https://data.austintexas.gov/resource/8jyt-x94k.csv and the cached comma separated values (CSV) file may be found under the Data/CourtData.csv sub-directory.

---

What Crimes Are Committed Most Often?

## Totals by crime (irregardless of date)
By.Crime <- d %>% group_by(Crime.Type) %>% 
  
  # Collapse data.frame by number of observations
  summarise(total = sum(Count)) %>% 
  
  # Add columns for cumulative distance and ranking
  mutate(cume_dist = cume_dist(total),
         rank      = dense_rank(total), 
         rank      = max(rank) + 1 - rank) %>% arrange(rank)

To kick things off, let’s take a look the frequency of each Crimes.Type under the APD Incident Report. To do this, we’ll aggregate each Crimes.Type into a data.frame using the summarise() function in dplyr.

This collapses our data.frame by an aggregate function, yeilding a new data.frame with two columns (the type of crime & the total number of overervations of that crime) and 129777 rows. For convience, I added two aditional columns to help me visualize and select some basic attributes by rank.

Distribution of Rank by Frequency

Now lets plot the ranking of each Crime.Type versus the number of occurances so far this year

Looks like it’s log distibuted. I’ll apply two transofrmations to the plot I just created to yeild:

That’s much better! Looks like there are quite a few crimes that occur with great frequency, so we’ll investigate those first.

Top 50 Crimes

Crime.Type	total	cume_dist	rank
CRASH/LEAVING THE SCENE	293874	1.0000000	1
THEFT	259057	0.9975728	2
BURGLARY OF VEHICLE	223076	0.9951456	3
FAMILY DISTURBANCE	217208	0.9927184	4
CRIMINAL TRESPASS NOTICE	87720	0.9902913	5
CRIMINAL MISCHIEF	84229	0.9878641	6
THEFT BY SHOPLIFTING	37916	0.9854369	7
DWI	36047	0.9830097	8
LOST PROP	35447	0.9805825	9
DISTURBANCE - OTHER	33626	0.9781553	10
ASSAULT W/INJURY-FAM/DATE VIOL	32052	0.9757282	11
HARASSMENT	24786	0.9733010	12
BURGLARY OF RESIDENCE	23622	0.9708738	13
WARRANT ARREST NON TRAFFIC	20215	0.9684466	14
PUBLIC INTOXICATION	20038	0.9660194	15
REQUEST TO APPREHEND	19464	0.9635922	16
ASSAULT WITH INJURY	17997	0.9611650	17
ABANDONED VEH	16024	0.9587379	18
AUTO THEFT	13840	0.9563107	19
POSS MARIJUANA	13798	0.9538835	20
FOUND PROPERTY	12201	0.9514563	21
FRAUD - OTHER	10454	0.9490291	22
CUSTODY ARREST TRAFFIC WARR	10288	0.9466019	23
IDENTITY THEFT	9865	0.9441748	24
EMERGENCY PROTECTIVE ORDER	9295	0.9417476	25
ASSAULT BY CONTACT	7334	0.9393204	26
DRIVING WHILE LICENSE INVALID	7111	0.9368932	27
POSS CONTROLLED SUB/NARCOTIC	7036	0.9344660	28
THEFT OF BICYCLE	6447	0.9320388	29
BURGLARY NON RESIDENCE	5494	0.9296117	30
CRIMINAL TRESPASS	4705	0.9271845	31
VIOL CITY ORDINANCE - OTHER	4546	0.9247573	32
DEBIT CARD ABUSE	3390	0.9223301	33
ASSAULT BY CONTACT FAM/DATING	3187	0.9199029	34
POSS OF DRUG PARAPHERNALIA	3069	0.9174757	35
THEFT OF SERVICE	3048	0.9150485	36
DATING DISTURBANCE	2790	0.9126214	37
BICYCLE REGISTRATION	2637	0.9101942	38
ASSAULT BY THREAT	2378	0.9077670	39
ASSIST COMPLAINANT	2272	0.9053398	40
BURGLARY INFORMATION	2050	0.9029126	41
CRED CARD ABUSE - OTHER	2010	0.9004854	42
TERRORISTIC THREAT	2007	0.8980583	43
DWI .15 BAC OR ABOVE	1780	0.8956311	44
THEFT INFORMATION	1701	0.8932039	45
DWI 2ND	1678	0.8907767	46
IMPOUNDED VEH	1440	0.8883495	47
ASSAULT INFORMATION	1423	0.8859223	48
AGG ASLT STRANGLE/SUFFOCATE	1331	0.8834951	49
FOUND CONTROLLED SUBSTANCE	1300	0.8810680	50

---

Incidents This Year

Now let’s look at the data set from the time perspective

## Totals by date (irregardless of crime)
By.Day <- d %>% group_by(Date) %>% 
  summarise(total = n()) %>% 
  filter(Date < as.Date("2015-11-01"))

Scatter Plot | Boring!

This is just plain black and white… So boring!

By Day of The Week

Scatter Plot | Colored by Day of Week

Scatter Plot + Linear Model | Colored by Day of Week

Line Plot Colored by Day of Week

It looks like Friday might be a bad day for crime. Let’s look at each day’s linear model, but include a scatter plot so we can get a better estimate

Scatter Plot + Colored Linear Models | Now we’re talking

Looks like we might be on to something with our Friday hypothesis. We can make a Violin Plot (a boxplot with the width representing density) to further investigate

Violin Plot | By Day of Week

Ah, that’s much easier to visualize.

This is still work in progress. Everything below this chunck is most 
certianly incomplete and will (hopefully) be finished at a later date!

Fig 04

Doesn’t look like anything special happens

GIFs

Contact

Hunter Ratliff

Email: HunterRatliff1@gmail.com
Twitter: @HunterRatliff1

Copyright (C) 2015 Hunter Ratliff

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <http://www.gnu.org/licenses/>.

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.