Module 1: Crime Analysis

-

Part C of the Module 1 assignment was to create hotspot analysis of homicides in the City of Chicago using three different methods. All three methods would utilize a point feature class of the total homicides for 2017 and the City of Chicago boundary. A second feature class of the total homicides for 2018 would then be used to see how well the hotspot predicted the location of homicides.

Grid-based thematic mapping

This method utilized feature classes for the total homicides for 2017, the City of Chicago boundary, and a ½ mile grid cell clipped to the Chicago boundary. All three feature classes were added to a map. The Spatial Join tool was used to create a new feature class that would contain the count of homicides for each grid cell. The Select By Attribute tool was used to select only the cells where the number of homicides was greater than 0. Next, the count column was sorted and the top 20% was selected and then exported to a new feature class. The final feature class was dissolved to create a single polygon for the hotspots.














Kernal Density

This method utilized feature classes for the total homicides for 2017 and the City of Chicago boundary. The Kernel Density tool was used to create a grid with cell size of 100 square miles.













Local Moran’s I

The last method utilized feature classes for the total homicides for 2017, the City of Chicago boundary, and the Census tracts polygons for the Chicago area. The Spatial Join tool was used to create a new feature class that includes the homicide counts in each of the census tract polygons. A new crime rate field was added to the attribute table and the rate was calculated using the below formula:

                !Join_Count! / !total_households! * 1000

The Cluster and Outlier Analysis tool was then used to determine the significant clusters of high crime. A SQL query was performed to select just the high-high clusters features. The selected features were exported to a new feature class and a dissolve was performed to create a single hotspot polygon.














The Kernel Density hotspot polygon contained 73% of the 2018 homicides but also the largest area. This would probably not be the best choice to present to the police chief. The Local Moran’s I hotspot polygon contained 45% of the 2018 homicides and would be my recommendation since it less area for the police force to cover.




Comments

Post a Comment

Popular posts from this blog

GIS Programming Module 1 – Python Environments & Flowcharts

-
Image
Module 1 examines “The Zen of Python” which is a poem by Tim Peters a software engineer. This poem can be seen by typing “import this” into any Python interpreter and describes what it means to write Python code. “The Zen of Python” is the principals or guidelines on how programmers should write code. These principles suggest to me that we should write code in a way that is easy to read and understand, even though the code may be complicated. I see this as writing code that is modular where the complexity can be broken down into smaller functions that do a specific task. Complexity can also be handled by simplifying the code, by finding ways to reduce the number of lines it takes to accomplish the same task. We should follow these principles not only for others who might later need to interpret the code, but for ourselves to help us in debugging and future alterations. It is up to the programmer to write code that provides the required functionality, but they should make a conscious ef...
Read more

GIS Programming Module 2 – Python Fundamentals

-
Image
  Module 2 assignment covered some of the fundamentals of Python. This included learning about the different Python data types, working with string variables, understanding loops and conditional statements, and iterating through variables.   The screenshot below shows the output of the script written for the assignment. Step 1 was to create a variable to hold my full name separated by spaces, split the variable using the spaces, and then print my last name using the index -1. The index -1 is the last item in the list which was my last name. Step 2 was to fix two errors in the code that was provided. The code uses a list of the first names of players and calculates the number of characters in players’ name. Each player rolls a dice which is randomly selected from a list of integers from 0 to 2 times the number of characters in the name. Players win if the dice is greater than the number of characters, tie if they are equal, and lose if the dice is less than the number of ...
Read more

GIS 5935 Lab 1 - Calculating Metrics for Spatial Data Quality

-
Image
The first lab of GIS 5939 concerned spatial data quality and being able to determine if the data is accurate, precise, or has any bias. This involved understanding the difference between accuracy and precision. The horizontal accuracy of a point is determined by how close the collected points are from the true location. This is calculated by measuring the distance from the average of all points collected and the reference point. The larger the distance means, the lower the accuracy. The horizontal precision is how close different measurements are from each other. Points that are clustered together have high precision and points that are spread-out have low precision. The image below illustrates accuracy and precision. Image Source: https://byjus.com/physics/accuracy-precision-measurement/ The map below shows red dots for waypoints that were captured using an unknown GPS device. These 50 waypoints were captured using the same GPS and at the same location. The points should be at the sam...
Read more