Investigating crime patterns in Egypt using crowdsourced data between 2011-2013

Dissertation submitted in partial fulfilment of the requirements for the degree of Master of Science in Geospatial TechnologiesCrime is a social phenomenon that negatively impinges upon the society on various levels. Such phenomena are ought to be measured and analyzed to achieve control over its presence and consequences. One of the ways for measurement and analysis involves the use of crime maps as vital tools for visualising crime related data. Getting access to crime data is undoubtedly a challenged endeavour faced by hurdles of data collection, storage and making it available for public access. In addition, coming up with useful relationships for extracting information and patterns for crime data analysis is a significant challenge as well. This research investigates the link between the spatial and temporal variables in crime related data collected from crowdsourcing. The research will capitalize on crime data gathered throughout the operation of an online project called Zabatak founded by the author since January 2011 in Egypt. The dataset consists of more than 2000 crime incidents from various geographical areas across Egypt. The research considers an exploratory analysis in trying to interpret crime patterns and trends. The results of this study have identified various interesting trends and patterns in the dataset. One of the major findings of this research points out a strong relationship between the spatial and temporal variables in Car-Theft incidents. In addition, It was possible in the study to relate crime types to the type of the geographical area. The research considers Spatio-Temporal analysis using Inhomogeneous Spatio-Temporal K-function and pair-correlation functions which have identified a Spatio-Temporal cluster and interaction in crime data which can open new ways for crime maps data analysis