Computer Science and Informatics
This REF submission reflects the core strengths of research in computer science and informatics that spans many departments in the University of Portsmouth. We returned three broad research themes: Computational Intelligence, Health Informatics and Networking and Security.
Key features of this submission:
- 100% of our impact was rated as either outstanding (4*) 40% or as being very considerable in terms of reach and significance (3*) 60%. We are rated first nationally for 4*/3* impact.
- Top overall world-leading research 4* amongst post-92 universities and University Alliance group.
- 20% of our research was rated world leading (4*) and 64% either world leading or internationally excellent.
- 19% of our research outputs were rated world leading (4*) and 60% either leading of internationally excellent.
- The average GPA increased by 48% from 1.9 in 2008 to 2.81 in 2014.
- Nationally, we have progressed from Q4 in 2008 to Q2 in 2014.
Research Groups/Research Themes
In Computational Intelligence, research is broad-based focusing on the design and implementation of systems that exhibit intelligent behaviour using a variety of sensor networks, methods, and data. Robotics and Computer Vision research comprises Approximate Computing, multi-sensor based information fusion, analytics and their practical applications, especially in cognition-driven biomechatronics, and pattern and face recognition and classification (Liu, Ju, Yu, Ait-Boudaoud, Chiverton, Jordanov). Research in Machine Learning (ML), Self-Organised systems, Optimisation and Neural Networks in particular, explore the hybrid meta-heuristic methods for global optimisation, based on Low-Discrepancy Sequences, Genetic Algorithms and Simplex search techniques (Jordanov, Ait-Boudaoud, Cocea, Wang, Frei). The development of computational intelligence methods with applications in modelling complex systems based on the newly pioneered research field of rule-based networks (Gegov).
Networking and Security
Networking and Security research falls into two interrelated areas: (i) security and digital forensics and (ii) mobile and wireless technology. We concentrate on the process of engineering secure distributed systems such as Grid and Cloud computing (Aziz, Liu), and more specifically on the applications of formal analysis techniques to the verification of security properties and modelling of systems in a secure manner (specification of security properties at an early stage of the software development process). We investigate through simulation the performance of networks for multithreaded architectures that include simulation of wireless networks to address performance of new protocols and quality of services for harsh environments (Adda). We engage in research on fusing information from novel sensing technologies, and we will concentrate on the advancement and development of algorithms that can fuse data from cameras, radars, lidars, bracelets, and unattended ground sensors, to yield a set of tracks that can identify anomalous behaviour (Adda, Aziz, Ju).
In Health Informatics, our collaborations with other academic institutions in the UK, US and Europe, health sector organisations (particularly Portsmouth Hospitals NHS Trust), and companies have enabled the development of research outcomes that were deployed in the health sector. We expanded our work on clinical outcome modelling to encompass further data sources, models specific to clinical conditions, and more sophisticated analysis techniques (Prytherch). We continue to facilitate the adoption of our models by the NHS and other organisations. In the area of telecare, we continue to seek large-scale data sources that provide us with raw data that we can use to model patterns of daily living.
Impact Case Studies
Improved mobility and quality of life for children with disabilities
Research at the University of Portsmouth has created new user-friendly control, navigation and communication systems for powered-wheelchairs, which benefit from effort-reduction, predictive and interactive Artificial Intelligence. These developments have made a significant impact on the lives of users, many of whom now have an opportunity for independent mobility, some for the first time. The systems have been used in six special schools and institutions (including RNIB and NHS) and many private homes. Economic impact in reducing the need for carers alone has been estimated at more than £250,000 p.a and the devices have also changed some professional services. The research was lead by the University of Portsmouth in collaboration with Chailey Heritage School and Sussex Community NHS Trust.
Clinical outcome modelling saves lives
Practitioner/professional service impact. Our work on clinical outcome modelling has influenced the Royal College of Physicians' (RCP) new standard for the assessment of the severity of acute illness (known as the "National Early Warning Score" or NEWS). The specific recommendation is for adoption by NHS bodies, but is already being adopted internationally.
Health impact. The chairman of the RCP working party estimated that our work could result in the saving of thousands of lives per year.
Economic impact. Our work is incorporated in the VitalPAC system developed by The Learning Clinic Ltd (TLC), and currently deployed to more than 20 hospitals.
Infrastructure and Facilities
In 2012, a strategic investment was made to create integrated Computing research and teaching facilities (infrastructure: ~£500K and equipment: ~£120K). A newly refurbished suite of laboratories which includes a digital forensics lab, a pervasive computing lab, a usability lab, and a mobile computing lab together with dedicated research spaces for PGR and PGT students have improved the research environment for students and staff. The creation of this dedicated space provides opportunities for researchers to engage in cross-thematic projects and it enables better interactions amongst PGR students and with PGT students. The success of this investment is evident in the increase of PGR recruitment in 2013, which has seen a three fold increase in 2 years (5 new PhDs in 2011 to 15 new PhDs in 2013). This new environment is also enabling effective support for early career staff to forge closer relationship with more experienced researchers.
All staff and researchers are provided with modern computing resources supported by a resilient University network infrastructure. In addition to providing departmental computing facilities and licenses for specialist software (e.g. MATLAB and others), our researchers use dedicated client software to access SCIAMA, a 1000+ cores distributed-memory High-Performance Computing cluster, managed by the Institute for Cosmology and Gravitation. (SCIAMA provides over 2TB of memory, 85TB of fast parallel storage, and 10TB of NFS storage, supported by 3 interconnection networks: 100bT, Gigabit, and Infiniband). A dedicated supercomputing technician supports these activities.