Results

The nearest neighbor fuzzy mode presents the frequency of violent crime incidents within a 3000 foot radius. Crimes are noticeably clustered around the downtown area and Peabody Darst Webbe, as well as the south-western part of the city, including neighborhoods such as Gravois Park and Dutchtown. The lowest frequency class is between 1-6 crimes, and the highest is 30-34 crimes.

A kernel density map was created for the study area to visualize the spatial distribution of violent crimes, and helps to visualize areas where there is a high intensity of crimes. The same neighborhoods are again highlighted as experiencing a high intensity of crimes, with the addition of Kingsway West & East in the north-eastern part of the city showing up as prominent areas.

An overlay hot spot analysis of the city was used to determine if independent variables were responsible in influencing the occurrence of crimes. The variables included poverty and bars, which are both known to be generators of crime. Under the theory of risky places, as outlined by Kennedy et al (2018), states that features of the physical environment act as places of conflict and generators of crime. Bars, due to attracting gang members and creating disorder due to the effects of alcohol, were therefore included. The results from the overlay hot spot analysis again were concentrated in the downtown areas of the city as well as the south-western neighborhoods. Public schools within a quarter mile of hot spot overlays were identified at potential risk to violent crimes. Overall, 16 public schools are at potential risk of experiencing a violent crime in these areas.

Grouping analysis was conducted to identify any prominent demographic features within each census tract. However no groups possessed any prominent variables, although the red groups did have a higher poverty rate and population, which may be a generator of crime. These red groups also occur within areas that have been previously identified as having a high amount of violent crimes.

Emerging hot spot analysis was used to create a unique understanding of the space-time relationship of crimes. The results from the emerging hot spot tool provide a mosaic of hot spots each with their own unique feature as explained below. One notable feature is that the consecutive hot spots are located within the centre of the overall hot spots generated, whereas the new hot spots are located on the edges, potentially a result of crime spreading outwards from existing hot spots.

New Hot Spot	A location that is a statistically significant hot spot for the final time step and has never been a statistically significant hot spot before.
Consecutive Hot Spot	A location with a single uninterrupted run of statistically significant hot spot bins in the final time-step intervals. The location has never been a statistically significant hot spot prior to the final hot spot run and less than ninety percent of all bins are statistically significant hot spots.
Sporadic Hot Spot	A location that is an on-again-off-again hot spot. Less than ninety percent of the time-step intervals have been statistically significant hot spots and none of the time-step intervals have been statistically significant cold spots.

Table 2: Definitions of the different types of hot spots (ESRI, 2019).