Citation
Clustering to Improve the Allocation of Drug Rehabilitation Resources

Material Information

Title:
Clustering to Improve the Allocation of Drug Rehabilitation Resources
Creator:
Ebben, Christina ( Author, Primary )
CU Denver Machine Learning Club
Publisher:
University of Colorado Denver
Data to Policy Project Symposium
Physical Description:
Research Poster

Notes

Abstract:
This research project focuses on how to allocate drug rehabilitation resources so that we can better aid those impacted by addiction. The Machine Learning Club has student members from a variety of academic fields, and we used our diverse skill sets to propose a way to implement this allocation strategy. Using Denver County crime data, we executed clustering algorithms using the geospatial location of drug arrests, concentrating on the most addictive drugs and specific factors that reflect a need for rehabilitation. Our results map where in Denver County the allocation of rehabilitation resources would be most effective and we include insight on what type of resources would be most beneficial in different parts of Denver. We conclude the project with suggestions of how clustering algorithms can be applied to other areas of the public sector.
Acquisition:
Collected for Auraria Institutional Repository by the Self-Submittal tool. Submitted by Christina Ebben.
General Note:
Data to Policy Project Symposium 2019

Record Information

Source Institution:
Auraria Institutional Repository
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

CU Denver Machine Learning Club Abstract This research project focuses on how to allocate drug rehabilitation resources so that we can better aid those impacted by addiction. The Machine Learning Club has student members from a variety of academic fields, and we used our diverse skill sets to propose a way to implement this allocation strategy. Using Denver County crime data, we executed clustering algorithms using the geospatial location of drug arrests, concentrating on the most addictive drugs and specific factors that reflect a need for rehabilitation. Our results map where in Denver County the allocation of rehabilitation resources would be most effective and we include insight on what type of resources would be most beneficial in different parts of Denver. We conclude the project with suggestions of how clustering algorithms can be applied to other areas of the public sector. Background What is Machine Learning? Unlike people, computers do not naturally consider past experiences when making decisions new data to make decisions based on what it has learned What is clustering? A type of machine learning algorithm where t he computer groups data points based on similarities Clustering Example The clustering algorithm clasifies Iris subspecies based on the length of the petal and the length of the sepal Methodology Decisions about the data Focus on the most addictive drugs in Denver County Focus on arrest types reflecting a need for rehabilitation Decisions about the machine learning We chose to implement clustering algorithms Decisions about analyze of the results Visualization and geospatial representation of results D ata Description The data was sourced from the Denver Open Data Catalog Filtered to the following (15,838 data points) Drug arrests from the year 2015 Barbiturates, Cocaine, Heroin, Methamphetamine, Opium, and Synthetic drugs Possession, Intent to Sell, and Manufacturing Latitude and Longitude of arrest Policy Recommendations Future Work References ArcGIS Desktop. Computer software. Redlands, CA: ESRI, 2010. City and County of Denver, Denver Police Department / Data Analysis Unit. Denver Open Data Catalog: Crime , (Data File). City and County of Denver, 2019. R Core Team. R: A language and environment for statistical computing. Computer software. R Foundation for Statistical Computi ng, Vienna, Austria: 2013. VanderPlas Mit Python: Das Handbuch Für Den Einsatz Von IPython , Jupyter , NumPy, Pandas, Matplotlib, Scikit Data.ipynb , 20 Nov. 2019, colab.research.google.com/drive/1r4Jk3cOIYtnIZ7NroOx4IVwwDdisY0rN#scrollTo=UpQzZS2PN_P6. Clustering to Improve the Allocation of Drug Rehabilitation Resources Clusters in the suburbs of Denver County would best benefit from advertising for the existing, private rehabilitation centers Clusters centered in the downtown area would benefit from additional K 9 police units and succinct patrolling routes average income neighborhoods would benefit from additional drug rehabilitation centers as well as counseling centers that offer harm reduction programs Further improvement for this research project Normalization of the data Consider overdoses in addition to arrests Additional ways that clustering algorithms could be used to improve the public sector Mental health and community counselling locations General patrolling routes Areas in need of affordable housing Geospatial Representation of the Data The data shows that drug arrests are centered downtown as well as an interesting disparity between drug arrests, with the majority being cocaine, heroin, and methamphetamine The results of the clustering algorithms map the geospatial locations of where drug rehabilitation resources would be most effective in Denver County (excluding the DIA region) Contributors: Christina Ebben, Evan Stene, Ian Arriaga MacKenzie , James Vance, Nick Koprowicz , Nick Mako, Shweta Yadav, Finian Blackett, Selvakumar Jayaraman, Swayanshu Pragnya