Data science
Our expertise spans the full data science life-cycle: from information management and privacy, via machine learning and representational logics, to practical applications in bio-health informatics.
Our facilities
We boast an incredible array of facilities, making our innovative data science research possible.
A key feature of our approach is closely coupling methodology and application. This creates a self-fulfilling loop, where challenging real-world problems drive the methodology research agenda, but also provide a natural route to exploiting new algorithms and methods.
We believe this deeply multidisciplinary approach is one of the distinctive features of data science at Manchester.
Areas of expertise
Our researchers focus their work in the following specialist areas:
-
Bio-health informatics
We are designing future technologies to revolutionise healthcare: from modelling biological processes to developing learning health systems, from low-energy sensors to harnessing multi-modal patient's data.
Contact: Robert Stevens
-
Data security and privacy
Data security and privacy is of paramount importance. People generate, process and share more personal data than ever before, and need protection. We analyse and design advanced cryptographic protocols that offer 'security-by-design' and 'privacy-by-design' in various real-world application areas, including e-banking, e-health, connected cars, smart grid and smart cities.
-
Description logics
Description logics are a well-known family of logics for knowledge representation that are used as logical underpinning of modern ontology languages, in particular OWL. We investigate various reasoning problems, including subjective ones like modularity, learning ontologies from data.
Contact: Uli Sattler
-
Machine learning
Machine learning is revolutionising the world. No other technology is progressing so rapidly in terms of real-world impact. We make world-leading contributions with new theory, methodologies and cross-discipline scientific impact, as well as consumer-driven applications.
Contact: Gavin Brown
-
Medical imaging
Modern technology (X-rays, CT scans etc) can look inside the body, but radiologists are overwhelmed with the huge variety and number of images that are being taken. We are developing software to measure organs, locate abnormalities and diagnose diseases.
Contact: Tim Cootes
-
Ontologies and metadata
Ontologies capture rich domain knowledge for intelligent applications and metadata. They enable a shared understanding of relevant terms in a domain, eg biology of healthcare between different users and computer applications. We are leaders in fundamentals, applications and more.
Contact: Uli Sattler
-
Text mining
The vast majority of information is in free text form. From theoretical modelling of language to large-scale processing of social media and literature to mitigating information overload and overlook, we are enabling computers to use natural languages like we do.
Contact: Goran Nenadic
Postgraduate research projects
Data science projects
- Automatic Activity Analysis, Detection and Recognition
- Automatic Emotion Detection, Analysis and Recognition
- Automatic Music Generation via Deep Learning
- Biologically-Plausible Continual Learning
- Blockchain-based Local Energy Markets
- Building Machine Learning Models Using Matrix Factorisation
- Contextualised Multimedia Information Retrieval via Representation Learning
- Data Integration & Exploration on Data Lakes
- Data Wrangling
- Deep Learning for Temporal Information Processing
- Design and Implementation of an FPGA-Accelerated Data Analytics Database
- Emerging Technology based Supply Chain Management
- Empirical Web Accessibility Evaluation for Blind Web Users
- Ensemble Strategies for Semi-Supervised, Unsupervised and Transfer Learning
- In between structure and texture: Strexture
- Information Component Analysis via Deep Learning
- Machine Learning and Cognitive Modelling Applied to Video Games
- Managing the data deluge for Big Data, Internet-of-Things, Industry 4.0 environments
- Minimum Description Length: An information-theoretic approach to the groupwise study of image information
- Multi-task Learning and Applications
- Problems in large graphs (e.g., social networks, VANETs)
- Scheduling, Resource Management and Decision Making for Cloud / Fog / Edge Computing
- Security and privacy in p2p electricity trading
- Social Human-Robot interaction using visual object recognition and hand pose estimation
- Specifying and Optimising Data Wrangling Tasks
- Zero-Shot Learning and Applications
Richard Banach projects
Ke Chen projects
- Automatic Activity Analysis, Detection and Recognition
- Automatic Emotion Detection, Analysis and Recognition
- Automatic Music Generation via Deep Learning
- Biologically-Plausible Continual Learning
- Contextualised Multimedia Information Retrieval via Representation Learning
- Deep Learning for Temporal Information Processing
- Ensemble Strategies for Semi-Supervised, Unsupervised and Transfer Learning
- Information Component Analysis via Deep Learning
- Machine Learning and Cognitive Modelling Applied to Video Games
- Multi-task Learning and Applications
- Zero-Shot Learning and Applications
Angelo Cangelosi projects
Andre Freitas projects
Aphrodite Galata projects
Simon Harper projects
Dirk Koch projects
Tingting Mu projects
Mustafa Mustafa projects
Norman Paton projects
Rizos Sakellariou projects
- Job and Task Scheduling and Resource Allocation on Parallel/Distributed machines
- Managing the data deluge for Big Data, Internet-of-Things, Industry 4.0 environments
- Problems in large graphs (e.g., social networks, VANETs)
- Scheduling, Resource Management and Decision Making for Cloud / Fog / Edge Computing