Applications in GIS (GIS5100): Lab 6 Crime Analysis

In this week's lab, we learned about different hotspot mapping techniques that could be applied to finding where certain crimes happening most often. The three techniques were the grid-based mapping, kernel density and Local Moran's I.

1.       When gathering the data for the top 20% of burglaries that were happening within the grid-based thematic map, I had 507 grid cells with different burglaries counts. I divided the 507 cells by 5 which gave me the number 101.4. I rounded down to 101 which I made my selection of the top 20% cells with the highest burglary count. The next map that created to pinpoint high burglary crime areas was a kernel density map. The parameters that I used for the kernel density analysis was 100ft for the output cell size and the search radius to be set at 2640 ft. After using the kernel density analysis tool, I reclassified the data and excluding any zero values. The mean number was 35.16 which I multiple it by 3 which resulted as 105.48. Any number below 105.48, I excluded from the map and convert it from raster to polygon. I had to use the select attribute tool to select features with the grid code of 1 to get the final resulting hotspot map. In the final hotspot map that was created, the Local Moran’s I was used. Before using the tool, the Blockgroups2009Fixed.shp and the 2007 burglaries file had to be spatial joined and a crime rate field had to be inserted into the data. The crime rate was determined by the number of burglaries per 1000 housing units. After the Local Moran’s I tool was used, the select attribute tool helped select the clusters with high crime rates.