Crime data at the national level is collated by National Crime Records Bureau & published as Crime in India annually. It is of great professional & research value & is used by stakeholders in public domain to prove or disprove their claims & also by Police depts. Some statistical models make it happen in a conventional way. This data is in the public domain & large part of it is digital. If used with Big Data mindset, it will provide analysis, opening vistas of predictive & preventive policing.
It can be done with Big Data Analytics tools. India tops road accidents is a matter of serious concern. The Hon’ble Supreme Court monitored the efforts of bringing down accidents, of Police agencies. If a data driven analysis of black spots is done with inbuilt causation & made available to the National Highways Authority of India & other road construction & maintenance agencies, it can be a small beginning for a big change. It’s like a user feedback guiding road engineering. It would get structured, documented and impact analysed.
Lots of other crimes / parameters can also be studied in a similar manner. Machine Learning with open source tools can lead us to Classification, Clustering & Predictions, way beyond our conventional mind can think. It is happening in some countries with immense success. Can we connect the traffic violation cases with incidence of road accidents? Late night driving & accidents? Method of issuing driving licences with accidents? Can we change driving behaviour based on these outputs? Can we design effective traffic awareness with the findings?
Every crime head, crime grouping & their correlation can be analysed by current Machine Learning & Artificial Intelligence tools. It can brought to dynamic & real time level as machines become competent to learn, analyse & predict. Palantir Gotham software could predict landmines in Afghanistan besides other counter terrorism analysis. It is used by US Intelligence Community & US Dept. of Defense.
LATENT CRIME DATA CAN BE WORKED UPON FOR BETTER POLICING.