Facilitating reusable datasets: a case study on locational data
The ability to reuse data could allow an organisation to use their resources more efficiently and deliver better value from the initial investment made.
Find out more
Facilitating reusable datasets: a case study on locational data
The ability to reuse data could allow an organisation to use their resources more efficiently and deliver better value from the initial investment made. Government and commercial organisations have accrued vast amounts of data over years of investment that is potentially worth billions of pounds but is often not used beyond its original purpose.
Most tracking data will come in the form of locational data collected via GPS devices (including mobile phones) or Bluetooth but tracking data could also be collected from cameras such as CCTV or drones. Analysis of this data could allow: the detection of suspicious behaviour in crowds; ensuring safety at events; understanding the effects of climate change on migration patterns; optimising transport networks or delivery routes; and improving the sustainability of supply-chains.
This project aims to understand the state of the practice through a review of current platforms that facilitate the reuse of locational datasets, resulting in a set of recommendations of how to promote data reuse and a set of characteristics of a reusable dataset. There are an increasing number of technologies that can track the movements of large groups (people or animals).
Enabling a culture of data sharing
This project will investigate the current state of practice with respect to data reuse by reviewing and comparing key sets of principles and guidelines that have been developed to facilitate data reuse.
Find out more
Enabling a culture of data sharing
The way data is designed, captured, and managed comprises approaches primarily intended for single use purposes resulting in huge amounts of data being generated. The collection of such data can be very costly; being able to share data within, and between, organisations could allow resources to go further and allow cost savings.
This project will investigate the current state of practice with respect to data reuse by reviewing, and comparing, key sets of principles and guidelines that have been developed to facilitate data reuse. A set of challenges that have not been addressed by the existing principles will be established that will form the basis of a future research agenda.
Attitudes to Data
This set of research activities explore the challenges facing contemporary organisations to provide confidential data that is available, when needed, and is trustworthy.
Find out more
Attitudes to Data
This set of research activities explore the challenges facing contemporary organisations to provide confidential data that is available, when needed, and is trustworthy. This is supports one of the primary objectives of the Defence Artificial Intelligence Strategy which is to transform Defence into an "AI-ready organisation” (2022:4). Existing research has suggested that one of the crucial ways to make organisations ‘AI ready’ is ensuring that data is secure, accessible, and reliable (Jöhnk, J., Weißert, M., & Wyrtki, K. 2021). As part of the DDRC, WP3 (Data and Desirable Digitalisation: Challenges, Security, and Ethical Considerations) this project explores two issues. First, UK residents’ attitudes to data in their organisations, as both workers and citizens. Second, understanding barriers within organisations to better use of data and being ‘AI ready’. These two areas overlap as we will explore both organisational and workforce understandings of barriers to better use of data in workplaces. Specifically, the research explores data flows, data security, and data privacy issues. The findings of the workshops and survey will provide guidance for best practice in managing processes such as data wrangling and optimising how organisations use data.
Data Management
Over recent years, experience with the use of digital data has grown to the point where a collection of mature practices has begun to emerge.
Find out more
Data Management
Over recent years, experience with the use of digital data has grown to the point where a collection of mature practices has begun to emerge. However, these are widely scattered across a variety of organizations and institutions. They also have uncertain applicability to the complex and demanding scenarios encountered across the MOD. As the use of AI and data-driven decision-making increases, the importance of efficient and effective data management approaches grows. Critical topics of concern will include:
- Current and emerging commercial technology capabilities for data management
- Usage patterns for data-driven decision making
- Best practices in gathering, organizing, and managing datasets in AI-based solutions
In particular, this project will focus on management, catalogue existing efforts in relation to MOD needs and, working with Dstl stakeholders, provide recommendations on how they could implement best practice.
In addition to baseline assessments and landscape studies, this project will aim to produce a review of available best practice guidance and a handbook of best practices for use in specified MOD scenarios.
Healthcare
This project will work with the Defence Medical Services to understand the demand for AI and data-driven solutions, identify the most beneficial opportunities for AI adoption and produce a roadmap for their implementations.
Find out more
Healthcare
Artificial Intelligence (AI) solutions have the potential to transform how healthcare is delivered by assisting in diagnostics, facilitating personalised treatment plans, assisting with triage, and improving efficiency in a patient’s healthcare journey. Understanding how to implement AI solutions to work alongside existing systems brings not only technical challenges but organisational ones too. These challenges can mean some areas are more suitable for AI deployment than others. This project will work with the Defence Medical Services to understand the demand for AI and data-driven solutions, identify the most beneficial opportunities for AI adoption and produce a roadmap for their implementations.
Synthetic Data
This project will review the state of the art in synthetic data generation. It will define metrics for assessing the quality of synthetic data and thus develop new algorithms for synthetic data generation.
Find out more
Synthetic data
Synthetic data is algorithmically generated data that mimics a natural data set. Synthetic data is important because it can be made publicly available for algorithm development without the risk of disclosing private information, thus allowing a greater range of researchers to work on a problem equivalent to the original. In addition, synthetic data can be used to augment scarce training data. This project will review the state of the art in synthetic data generation. It will define metrics for assessing the quality of synthetic data and thus develop new algorithms for synthetic data generation. The project will concentrate on the relatively unexplored areas of coherent data, such as audio signals, and on discrete data which arises in many applications.
Education
The DDRC Education programme focuses on increasing the AI and data literacy of stakeholders within the Defence sector.
Find out more
Education
The DDRC Education programme focuses on increasing the AI and data literacy of stakeholders within the Defence sector. The programme includes a variety of different sessions that focus on demystifying and building business understanding about data-driven disruption, emerging technologies and AI – using clear, non-technical language to maximise accessibility.
Resilience
Data resilience protects critical info, ensuring availability despite disruptions, essential for maintaining organizational reputation and customer trust.
Find out more
Resilience
Data resilience ensures critical information is available regardless of any disruptions or data loss. The lack of data resilience strategy can severely damage an organisation's reputation and impact the confidence of both customers and partners.
This project is focused on approaches to data resilience. The project will be focusing on two research areas: (1) techniques and mechanisms for increasing trust and confidence in data, with a focus on evidence of tampering, corruption, alteration, and loss; and (2) mechanisms and architectures for making organisations resilient to losses, disruptions and unavailability of data - with particular attention on architectures, decentralisation, duplication, redundancy, recovery.