Fuzzy Based Geo-Spatial Crime Category Prediction for Crime Mapping and Safe Route Travel

Crimes are rapidly increasing in complexity and technological advancements across the globe. Prioritizing action to target specific crime categories within any area has become imperative. Moreover, civilians should be more aware of crime hotspots while travelling, living in, or entering unfamiliar places. Such an awareness ensures that timely action can be taken if and when any crime incident occurs. As a potential solution to this problem, this work proposes a combination of a Safe Route Travel app and a Crime Mapping system that utilizes a crime prediction model based on the Fuzzy K-Nearest Neighbour algorithm and geo-spatial operations. The model predicts the type of crime to which a particular place is most vulnerable based on historical crime data for that place. For implementing the model, an extensive dataset consisting of over 80 crime categories and ample crime records since 2016 has been used. However, this model can be utilized for any location where geo-encoded crime data are available. The overall system plays a vital role in reducing the delay between the police and the commoners. The Crime Mapping system can assist police in mitigating crime rates by helping them understand and study crime patterns, while the Safe Route Travel app can help civilians stay vigilant.