I am interested in the application of data science, computer vision and data visualisation to the Crime Science domain with a particular focus on policing. The main focus of my research is combatting future crime; specialising in detecting and predicting emerging crime trends as well as future methods of perpetration.
I also have a few small feats I wish to overcome, namely:
- Changing public perspectives on Data Science and the use of Data to aid law enforcement.
- Encouraging and promoting more efficient data recording, sharing and analysis (through a single system) at law enforcement establishments.
- Developing and equipping a new wave of crime scientists with an in-depth understanding and knowledge of data science.
My current studies and research, with much appreciation and gratitude, are being funded by the Engineering and Physical Sciences Research Council (EPSRC) through the UCL Dawes Centre for Future Crime.
PhD Topic: Horizon Scanning Through Computer-Automated Information Prioritisation
It is no secret that police forces are seeking to advance their repertoire of analytical and predictive tools to deal with emerging crimes as they adapt to the repercussions of penurious fiscal policies. However, the volume, variety, and velocity of data recorded by police forces and available through open sources means that analysing such data can be an ambitious, and challenging undertaking. Despite these obstacles, horizon scanning (HS) is set to become a prominent aspect of future policing, by enabling police forces to identify emerging threats, anticipate imminent crimes and to ultimately extinguish future methods of perpetration. In essence, its purpose is to aid the police in staying one step ahead of criminals.
This research aims to develop a novel automated technology that can identify early warning signals of future change. This will be split into three main parts, utilising what the police know, what the community knows, and what the media knows. Each part will build on techniques from the previous, to create a single, comprehensive, HS scanning system. In order to develop research output that is relevant, usable, and practical in the real world, this research will foster close working relationships with law enforcement agencies to resolve current blockades in HS. This will be achieved by developing prior work on the issues faced by practitioners (strategic police analysts) in the HS field.
Firstly, a conceptual piece will be undertaken to help identify and define the constitutional motifs of the research such as the characterisation of new crime types. This will then inform subsequent work on automatically detecting and monitoring crime trends.
A framework for an open-source intelligence repository (OSINTR) will then be created to automatically collect, and store, a large quantity of data from different sources. The repository will be initialised with ample sources for this research, including; news reports, social media data and community forum posts. However, the framework will be fully expandable to incorporate additional data. This will formulate the basis of subsequent work in conjunction with police data.
The first part of the research will develop a crowdsourced dataset for object and action detection in textual data since these are anticipated to be key elements in understanding new crimes types. The dataset will then be used to evaluate various natural language processing techniques for extracting descriptions of the actions and behaviours exhibited by an offender whilst committing a crime. This will then be replicated on crime reports to establish performance in this context and facilitate recommendations for police data collection practices. Anomaly detection techniques (similar to those used by banks to identify fraudulent transactions) will then be implemented to test the extracted information’s potential for crime monitoring. Further work may also be done on analysing the spatial and temporal spread of new types of crime.
The second part of the research will utilise the OSINTR, deep learning (where computers learn from data, without being explicitly programmed), and different datatypes to further the methods used in part one for theme extraction and detection. This will be followed by an investigation into the accuracy and interpretability trade-off of using black-box approaches.
The research will then conclude with a final study on predicting tomorrows crime headlines today by utilising the different data sources and employing self-learning algorithms. The purpose of this is to generate short problem snippets that analysts can then investigate further.
The primary objectives of the research are to develop an automated HS system that:
- Automatically collates information from a variety of open data sources
- Can extract crime related information from different data types
- Monitors anomalous changes in modus operandi until they become problematic
- Can make predictions in the form of future news headlines
Click here to view my profile on the UCL Dawes Centre for Future Crime Website as well as other PhD research occurring within the centre.
Here is a list of research projects that I am currently participating in.