Better Data in Forensic Science
What are we doing?
Forensic science research data has been generated for decades, both in forensic science laboratories and by academic researchers, but much of the raw data is inaccessible or lost. Traditionally within forensic science there has been little or no culture of sharing or preservation of research data and what data exists is spread across many sources with little organization, centralization or accessibility.
Part of the vision of the Leverhulme Research Centre for Forensic Science (LRCFS) is to provide the leadership, organization and funding needed to develop centralized data pools derived from existing data resources. As far as possible, these data pools will be made openly accessible and freely available across the practitioner and researcher communities to the benefit of all who work in the service of the justice system.
Making the most of data is vitally important. This is achieved through improved data handling techniques made available through Data Science and Data Visualisation methods. We are working across the LRCFS team and with forensic science providers, industry leaders and academic groups to develop better awareness and best practice in working with data.
Why are we doing it?
Data informs the majority of decisions in people’s lives. Forensic Science is no different. Ground truth data is the bedrock of some evidence types. We can assign robust statistical probabilities to DNA profiles only because we have data to show how common each aspect of the profile is in the general population. This is not possible for many evidence types and there is a need for more data in order to address this challenge.
How data is used is equally important, as misrepresentation of information can lead to incorrect inferences and poor decisions. Creating open, shareable research data together with the statistical and visual tools for working with it will enable better interpretation of data in Forensic Science. The generation of open source data and validated data handling and statistical tools is a core aim of LRCFS.
How will we do it?Citizen Science, Hackathon, Strategic Conversation
Across all our projects we are ensuring that we fulfil the FAIR data principles meaning that any data we produce as far as possible, will be findable, accessible, interoperable and re-useable.
We will develop freely available, open source toolkits and publish guidance on how to work with forensic data for the benefit of researchers and frontline caseworkers.
DataFest19 Hackathon event and data release
The LRCFS Datafest19 Hackathon
Data released as part of the Hackathon event: