Skip to content

New York crime analysis - R - Data mining course - association rules - density clustering(DBSCAN) - hotspots detection - mapping crimes

Notifications You must be signed in to change notification settings

mcoric96/Crime-analysis

Repository files navigation

Crime-analysis

For the best view (html dependencies - datatable and leaflet libraries) please download Crime_analysis_project.html file and have a look in your browser.
General purpose of this project is exatraction of knowledge related to crimes in this data and finding important feautures of crime in New York.

New York crime data

date/time of crime
Latitude/Longitude
NewYork borough
location
crime description
offense level
premise description

Kaggle original data

www.kaggle.com/adamschroeder/crimes-new-york-city/version/1#

R programming language(3.6.0)

See Crime_analysis.md

Data mining course - Mathematics and computer science department

Data visualization
Basic statistics
Association rules - support, confidence, lift, Apriori algorithm
Hotspots detection - DBSCAN clustering
Nearest neighbor index - z test statistics

Preview

Distribution of crimes in different boroughs


Crime events


Association rules


Vehicle crimes - mapping crimes

About

New York crime analysis - R - Data mining course - association rules - density clustering(DBSCAN) - hotspots detection - mapping crimes

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages