The rise of open data coupled with increasingly powerful methods in data science and artificial intelligence offers unparalleled opportunities to gain insight into, and make predictions of, human criminal behaviour. Simultaneously, we have a collective responsibility to ensure that data and methods are used in a responsible manner. Data-driven justice aims to harness the power of big data and cutting-edge data science technologies to find ways to maximize fairness, reduce criminality, and improve the delivery of justice.
The challenge is to find and use open data to inform community groups and law enforcement within Europe. For instance, we could explore optimal allocation of policing resources to make high crime areas safer for residents. Alternatively, we might explore whether bias in policing practice leads to disproportionate incarceration of minority populations. Throughout the challenge, participants must make explicit effort to identify and potentially address social, legal and ethical issues related to the administration of justice.
Note: If you intend to learn the hard way (preferred method)then we'd strongly advice to write as much code as you can yourself and not just run pre-written code. If you still want to test it, we can also share notebooks via Nbviewer
- Command Line Tool
- Anaconda with Jupyther Notebook
- Git intstalled, Github account
There's many options to have python, we're going to use Anaconda
Anaconda is a free distribution of the Python programming language for large-scale data processing, predictive analytics, and scientific computing that aims to simplify package management and deployment.
Follow instructions to install Anaconda or the more lightweight miniconda.
Is highly recomended to download Tableau Public and make an account to share data stories.
You can also apply for an Academic licence for Tableau Desktop.
Example: Pedro V
-
General Tutorials
https://github.com/donnemartin/dev-setup
https://github.com/christophebourguignat/notebooks
https://github.com/sindresorhus/awesome
https://docs.python.org/3/library/json.html -
Documentation
http://www.systemicpeace.org/index.html
http://jsoneditoronline.org/
https://www.theverge.com/2018/1/17/16902016/ compas-algorithm-sentencing-court-accuracy-problem -
Datasets
https://github.com/caesar0301/awesome-public-datasets
https://dataverse.harvard.edu/
https://www.opendatanetwork.com/
http://esango.un.org/civilsociety/login.do
http://www.icpsr.umich.edu/icpsrweb/NACJD/discover-data.jsp
http://www.europeansocialsurvey.org/data/
https://www.gdeltproject.org/data.html
https://data.nlc.org/dataset/APD-Crime-Summary/tzxe-h2tb
https://data.nlc.org/dataset/2014-SOTC-Database-real-/fn6z-uuyx
https://github.com/emorisse/FBI-Hate-Crime-Statistics/tree/master/2013
https://knoema.es/atlas/Holanda/datasets
http://www.icpsr.umich.edu/icpsrweb/NACJD/studies?q=&restrictionType[0]=Public%20Use&classification[0]=NACJD.XXV.*&dataFormat[0]=Delimited
https://www.nscr.nl/en/dataset-criminal-career-life-course-study-beschikbaar/
https://www.nscr.nl/en/datasets/