Crime Analysis Mod01

The first week's lab focused on using different techniques in crime analysis. Three hotspot maps of 2017 homicides in the Chicago area were created. These maps were then compared to the 2018 homicides in the Chicago area to determine which map technique was the best for predicting these crimes in the future.  

The grid-based thematic hotspot mapping where the 2017 homicides and the homicide grid were spatially joined. The top 20% of grid cells were manually selected and exported to a new feature class. 




The kernel density hotspot mapping where the kernel density tool was used and the data was reclassified to the mean values and the data was exported to a new feature class. 


The local Moran’s I hotspot mapping where the cluster and outlier analysis tool was used and the high-high clusters were dissolved.




I believe that the kernel density hotspot map is best for predicting future homicides as well as being the best representation of the information. The kernel density provides smoothing to the data giving a better idea of the areas where they are occurring. The other hotspot maps are squared off not conforming to the shape of the actual areas and just providing a general area.


Comments

Popular posts from this blog

Intro to GIS - Lab1 Overview of ArcGIS

About me

Visibility Analysis Mod03