Background Information
Background Information
Chicago police in order to keep track of crimes taking place in its region utilises Citizen Law Enforcement Analysis and Reporting (CLEAR). For the present task information of crimes is collected from 2012 to 2016 (Kieltyka, Kucybala, & Crandall, 2016). Information is provided on different crimes which have taken place segregated by location and location description, domestic violence, arrest, beat and district.
Dataset1 was selected to investigate crimes in Chicago. The information pertains to a period of 2012 to 2016.
From the analysis of the information it is found that the 361741 crimes took place between 2012 to 2016. The data pertains to 32 different types of crimes. Crimes in Chicago are classified into 32 different categories. Moreover, it is found that 544 crimes are the maximum number of crime which took place at a particular location (Kennedy et al., 2016). The least number of crime which took place at the location is 152 (Drawve & Barnum, 2018).
Investigation into the dataset highlighted the fact that the three highest frequency of crimes were Theft, Battery and Narcotics. The least three frequency of crimes were non-criminal, human trafficking and non-criminal. Analysis shows that the highest frequency of crimes took place in 2012. The least number of crimes took place in 2016. In the month of December 19 and 31 crimes took place in 2015 and 2016 respectively.
The year 2012 reported the highest number of crimes in Chicago.
The trend of crimes in different is approximately the same. It can be viewed from the trend that most crimes take place in January. Moreover, most of the crimes does not lead to arrest. Further, it is found that the dataset pertains to information of 23 districts of Chicago. In addition, most of the crimes took place in district 11 (25.56K).
The least number of crimes took place in the district of 31. Theft reported the highest number of crimes in district “8” in 2014. During the period of 2012-16 there were 53161 domestic crimes. In 2012 there were 132.72K domestic crimes. Similarly, the number of domestic crimes in 2013 was 121.13K. Further, it is found that there were 106.14K crimes in 2014.
For the advanced insight we investigate the crimes committed in 2016.
From the above image it is found that in 2016 theft had the highest frequency of crimes committed in 2016. The least crime committed was homicide.
Reporting/Dashboards
The above image presents the crimes in different months of 2016. It can be envisaged the frequency of crimes in April and May was higher than in other months. It can also be foreseen that Gambling and Stalking was the only crime in April.
Further investigation of crimes in May showed that most crimes took place in Thursday. It was also seen that no crime took place on Saturdays and Sundays in the month of May in 2016.
Further investigation of crimes in April showed that most crimes took place in Tuesday. It was also seen that no crime took place on Wednesday, Thursday and Saturdays in the month of April in 2016.
The above image presents domestic violence in different months of 2016. It can be seen from the above image that most domestic violence took place April and May.
The analysis of the information suggests that there are many locations where only one crime has taken place. In addition, we find that the highest crime took place at 41.75, 87.74. The most crime at a location was 69. Moreover, it is found that the number of crimes on each day of the week was approximately equal.
In order to analyse the data different visualization techniques were used.
In order to depict the number of different crimes in 2016 a bar chart is used. The bar chart is found suitable since the height of the chart is proportional to the number of crimes of the particular type. Thus, a comparative study of the frequency of different crimes can be done (Evergreen, 2016).
The frequency of crimes every month is depicted through the use of a line chart. A line chart has an x-axis and y-axis. On the x-axis the months are placed and in the y-axis numbers are provided. The placement of the legend relating the x-axis and y-axis provides information for crimes every month (Miller, 2016).
In order to visualise the highest number of crimes for a location a cross table is used. The row axis represents the location and the column axis presents the number of crimes. The column axis is sorted from highest to lowest. Thus, one can easily view the highest number of crimes at the particular location (Song et al., 2017).
Recommendation 1: Location of the Crime
From the above visualisation the police chief can easily contemplate that the frequency of crimes is higher in the districts of 11, 8 and 4. There seems to be an influence of neighbourhood (district location) in crimes. Thus it can be recommended that the police chief should be highly vigilant in these districts (Schnell, Braga & Piza, 2017).
Advanced Insights
Recommendation 2: Type of the Crime
Further from the analysis it is found that the highest frequency of crimes over the last 4 years is theft. Thus the department needs to be extra-vigilant towards theft crimes.
From the above analysis it can be envisioned that the incidence of crime is higher in district 11, 8 and 4. thus it can be suggested that the police department should particularly focus on these districts (Yang et al., 2018).
Further it is also found that the frequency of crimes in January is higher than in other months. Thus it can be recommended that the department needs to increase its vigilance during January as compared to other months.
The use of an online BI tool to investigate crime data was challenging. IBM Watson Analytics was easy in the sense that it uses AI to generate approximate visualisations. For each of the analysis a suitable question had to be asked. Initially we had to visualise how to represent the answers to the questions. The challenge was to ask the particular questions which would provide the given answer. The next task was to select the particular suggested visualization which would be easy to understand and would provide the necessary answer.
Reference
Drawve, G., & Barnum, J. D. (2018). Place-based risk factors for aggravated assault across police divisions in Little Rock, Arkansas. Journal of Crime and Justice, 41(2), 173-192.
Evergreen, S. D. (2016). Effective data visualization: The right chart for the right data. Sage Publications.
Kennedy, L. W., Caplan, J. M., Piza, E. L., & Buccine-Schraeder, H. (2016). Vulnerability and exposure to crime: Applying risk terrain modeling to the study of assault in Chicago. Applied Spatial Analysis and Policy, 9(4), 529-548.
Kieltyka, J., Kucybala, K., & Crandall, M. (2016). Ecologic factors relating to firearm injuries and gun violence in Chicago. Journal of forensic and legal medicine, 37, 87-90.
Miller, J. D. (2016). Learning IBM Watson Analytics. Packt Publishing Ltd.
Schnell, C., Braga, A. A., & Piza, E. L. (2017). The influence of community areas, neighborhood clusters, and street segments on the spatial variability of violent crime in Chicago. Journal of quantitative criminology, 33(3), 469-496.
Song, J., Andresen, M. A., Brantingham, P. L., & Spicer, V. (2017). Crime on the edges: Patterns of crime and land use change. Cartography and Geographic Information Science, 44(1), 51-61.
Yang, D., Heaney, T., Tonon, A., Wang, L., & Cudré-Mauroux, P. (2018). CrimeTelescope: crime hotspot prediction based on urban and social media data fusion. World Wide Web, 21(5), 1323-1347.