Data&Codes.zip

This project is linked to a manuscript submited to International Journal of Geographical Information Science: A statistical framework for measuring local bivariate spatial association using semantic indicators and an exact parametric method.This file contains data and codes. The data contains two real-world datasets. The first is the crimes data for Philadelphia in 2022, which is publicly accessible and stored in the directory 'Data/Ph_crimes_data'. The second is the traffic crash data for Hong Kong spanning the years 2014 to 2019. This traffic crash dataset is not publicly available, so only a portion of it is provided for reference purposes, and it is stored in the directory 'Data/HK_crashes_data'. The detailed description could see in 'README.me'.The code consists primarily of two components. The first and main component is the '<b><i>Data_to_result.ipynb</i></b>' file. This file can be opened with any text editor and provides a detailed summary of the entire workflow described in the manuscript for analyzing Philadelphia crime data. It outlines the process from reading the raw data, calculating local indicators, generating plots, to constructing tables. By executing this file step by step, one can gain an in-depth understanding of each stage of the work-flow.To facilitate the widespread use of the proposed framework, we have implemented it as the <b><i>ESLocalIndi</i></b> open-source package in Python, making it easily accessible to geographers. The framework has been uploaded to GitHub. To prevent the disclosure of sensitive information, such as the authors' affiliations, during the review process, we have set the project status to private. It will be made publicly accessible once the review is complete. The source code of the proposed framework stored in the local directory '<b>ESLocalIndi</b>'.