https://github.com/MechaKaradi/163B-Repository-Group-4.git
Authors: Lala Sayyida Millati Nadhira 5844266 Kaninik Baradi 5216664 Rezzy Yolanda Wulandhari 4779487 Kelvin Engee 4664043 Philippe Almeida Mirault 5898803
The relationship between green spaces and crime rates is debated, with some studies suggesting that greenery can have a calming effect and attract foot traffic to deter criminal activity (Kuo & Sullivan, 2001). However, the relationship is not always clear-cut and may be influenced by various factors. In certain contexts, green spaces may even increase crime rates by providing cover for criminal activity or attracting individuals more likely to engage in crime. Understanding the factors that may have contributed to the crime would be helpful in predicting the likelihood of future crimes occurring.
This repository handles the import, cleaning, and visualisaton of data from the NYC Open Data repository to develop the descriptive and predictive analytics for the model.
The environment.yml file provides the list of conda dependencies required to run the notebooks.
Relative to the root folder of the repository, the following files are needed:
..\\NYPD_Complaint_Data_Historic.csv: source..\\data\\2015 Street Tree Census - Tree Data.geojson: source..\\data\\Police Precincts.geojson: source
The Notebooks are numbered in the order they should be run. The function of each notebook is described below.
01_Data Clean Final.ipynb : Cleans the Crimes data from the NYPD_Complaint_Data_Historic.csv and generates a pickle file with the cleaned data. This file is used as input for the other notebooks.
11_CrimeClusters.ipynb : Prepares the data for the machine learning models. It generates a grid, counts the crimes near the grid points, and generates a pickle file with the data. This file is used as input for the other notebooks.
12_YearsSeperated.ipynb : Performs the same pre-processing, but seperates data from 2021 for validation.
21_RegressionOnTrees : Performs the initial regression models using the location, time, and trees information,
22_RegressionValidation : Performs the same regression, but also validates it using data from 2021
The count_trees.ipynb notebook file generates reduced analytics for the number of trees in each precinct.
The remaining notebooks generate visualisations and analytics.