As part of our partnership with the Alan Turing Institute, the Institute of Advanced Data Science hosts six flagship university projects. They sit within a wider landscape of funded projects undertaken at the Institute of Advanced Data Science. Further information about our research can be found on our members' webpages.
The Alan Turing Institute - Queen Mary IADS Flagship Research Projects
Professors Conrad Bessant and Trevor Graham are Principal Investigators on the project: 'Automatic Learning In Cancer Research Using Artificial Intelligence'. The goal of the project is to create and apply an automatic research methodology to rationally design mechanistic mathematical models of cancer evolution, and then evaluate the models using public and private data repositories. The project will establish the value of artificial intelligence approaches in basic biology, and make further inroads into understanding the large and complex datasets available in cancer research.
Dr William Marsh is Principal Investigator on the project: 'Knowledge Discovery from Health Use Data' (KNIFE). The goal of the project is to develop techniques to extract actionable knowledge from Electronic Health Records. The focus will be to establish data-driven methodologies for supporting a number of key medical challenges, including discovering of illness causes, optimising health delivery and choosing treatment. The project team includes professors Norman Fenton & Martin Neil, Dr John Robson and Research Assistant Haoyuan Zhang. Project partners include the Clinical Commissioning Group of Tower Hamlets NHS.
Professor Andrea Cavallaro is Principal Investigator on the project: 'Privacy-preserving Multimodal Learning for Activity Recognition'. The goal of the project is to design and validate efficient on-device models, built from multimodal sensor data from body cameras. The focus will be to develop privacy-preserving aggregation, processing and interpretation algorithms to enable us to deepen the scientific understanding of the core problems in the application domain of body cameras.
Professor Sean Gong is Principal Investigator on the project: 'Deep Learning for Large-Scale Video Semantic Search'. The goal of the project is to develop algorithms for fast search and summarisation of very large quantities of unprocessed videos data. The focus will be on analysing large-scale video data (e.g. CCTV footage) for the estimation of population movement, and searching for categories of people in urban environments. This will enable application such as crime prevention and urban planning for safer transport and roads in emerging smart cities. Project partners include Nvidia, IBM, Huawei Europe, OCF, SCC, SeeQuestor and Vision Semantics.
Professor Kaspar Althoefer is Principal Investigator on the project: 'Learning collaboration affordances for intuitive human-robot interaction'. The goal is to create intelligent methods for natural and intuitive human-robot interaction, with the goal of equipping robots with the required intelligence to understand a given manual task in such a way to actively support a human. The focus will be on "tool handover" in work environments, such as an operating theatre or nuclear power plant. The project will enable robots to better understand complex multimodal cues used by humans when tools are handed between people and robots. Project partners include the National Centre for Nuclear Robotics, Innovate UK and SBRI funds for higher TRL industrialisation.
Professor Steve Uhlig is Principal Investigator on the project: 'Learning-based reactive Internet Engineering' (LIME). The goal of the project is to design reactive communication networks, by relying on distributed Internet monitoring and data collection. The focus is to build feedback algorithms to automatically adapt network management practices, towards better usage of network resources and more secure network infrastructures. The project will enable faster and more secure Internet infrastructures. Project partners include GEANT, NEC & Huawei, and the team includes Dr. Gianni Antichi and Dr. Felix Cuadrado.
The Alan Turing Institute - Queen Mary IADS Research Projects
Living with machines project is a five-year research project that will take a fresh look at the well-known history of the Industrial Revolution using data-driven approaches. Initial research plans involve scientists from The Alan Turing Institute collaborating with curators and researchers to build new software to analyse data drawn initially from millions of pages of out-of-copyright newspaper collections from within the archive in the British Library’s National Newspaper Building, and from other digitised historical collections, most notably government collected data, such as the census and registration of births, marriages and deaths. The resulting new research methods will allow computational linguists and historians to track societal and cultural change in new ways during this transformative period in British history. The project is led by Professor Ruth Ahnert and colleagues from British Library, University of Exeter, University of East Anglia and the University of Cambridge.
Citizen participation and machine learning for a better democracy project aims to use machine learning and natural language processing to overcome information overload in citizen participation platforms. The project aims to address barriers to achieving effective direct democratic systems. If successful, it will allow citizens to directly contribute to the most important decisions in their communities. The aim is to test the hypothesis that the use of machine learning and NLP techniques on a digital platform for citizen participation will significantly increase the capacity of citizens to participate in local democracy. The project is led by Professor Rob Procter, Miguel Arana Catania, Professor Maria Liakata and Dr Arkaitz Zubiaga.
Towards incoherent speech as a predictor of psychosis risk project aims to assess which of the existing methods for characterising individuals’ speech is most likely to be predictive of psychosis in subjects at clinical high risk of psychotic disorders. The project will also develop new data science approaches to measure patients’ speech, and investigate whether these new methods provide extra power to detect psychotic disorders. It is led by Professor Maria Liakata, Professor Ginestra Bianconi and colleagues from the University of Cambridge.
AI for precision mental health project aims to produce AI tools that can personalise mental health profiles and advance the precision of early diagnosis and subsequent treatment. The aim is to develop predictive models based on machine learning approaches, to differentiate asymptomatic populations (those not presenting symptoms) at high versus low risk of mental health related disease. These models will be used to interrogate the neurocognitive factors that underlie cognitive health. The work has so far focused on dementia and has developed predictive and prognostic machines that are trained and validated to reliably predict individual rate of cognitive decline at early dementia stages from low-cost, less invasive data (e.g. cognitive testing, MRI scans) (update: September 2019). The project was led by Professor Zoe Kourtzi and Professor Maria Liakata.
Raphtory: a practical system for the analysis of dynamic graphs project aims to develop an open source system for dynamic graph analysis of datasets. Dynamic graphs allow for studying of how relationships form and change over time. Although they have many applications, there are no readily available tools and systems enabling their application. Raphtory is a new distributed system that enables dynamic graph analysis starting from very large real-time datasets. This project will improve the functionality and usability of Raphtory so that it can be readily usable by domain-specific researchers. The project will develop a set of use cases for dynamic graphs, starting with urban analytics for mobility incentives. This project is led by Dr Felix Cuadrado, Dr Richard Clegg and Professor Susan Grant-Muller.
Risk and uncertainty in peatland carbon emissions project aims to develop a framework to integrate different types of information of peatlands: quantitative data on greenhouse gas fluxes and environmental drivers such as temperature and precipitation; semi-quantitative understanding of controls and interactions, such as feedbacks between peat formation and water loss; and qualitative information on socio-economic factors, such as land-use policy and site-level management actions. The project is led by: Professor Lisa Belyea, Dr Peter Levy, Dr Emily Lines and Professor Norman Fenton.
Geometry and topology of complex interconnected systems is a project at the intersection of geometry, topology, algebra, and probability. The goal is to answer the following questions: how probabilistic effects appear at different scales in complex interconnected systems, how an understanding can be used to verify hypotheses about different complicated datasets, and how prediction and statistical analysis can take these structures into account. This project is led by Professor Michael Farber and part of the data-centric engineering programme's 'Mathematical foundations' challenge.
AI for control problems project uses a competition platform to accelerate progress in data-driven control problems. 'Rangl' is a competition platform created at the Alan Turing Institute as a new model of collaboration between academia and industry. Through integration with OpenAI Gym, rangl offers a user-friendly environment to develop learning approaches to data-driven control problems. Anybody can propose a rangl challenge, compete in a challenge by designing a controller, or contribute an 'off-the-shelf' AI controller for users to customise. This project is led by Professor John Moriarty.
Solar nowcasting with machine vision project will establish a worldwide open data “clearing house” for solar photovoltaic geodata. The clearing house will reconcile data sources, and transform the data into clean datasets consumable directly by machine learning algorithms. The algorithms being used for PV out-turn prediction by regional and national network operators (National Grid) and commercial market participants. This “missing link” will be a force multiplier, reducing carbon missions by enabling new demand management and energy-trading innovations. This project is led by Dr Dan Stowell and colleagues from University of Sheffield, University of Oxford and Open Climate Fix.
Detecting and understanding harmful content online project aims to systematise research in online harms (e.g. research on hate speech). It will do this by developing lists of datasets and benchmarks to compare different attempts to solve the problem and developing guidelines for practitioners who wish to use the outputs of online harms research. The project is led by Professor Nishanth Sastry, Dr Bertie Vidgen and Dr Gareth Tyson. A result of this project includes a hatespeechdata.com website which collects together several known datasets related to hate speech online. This is expected to serve as a resource for researchers in the area.
Security and privacy in the decentralised web project is looking at the emerging decentralised web - specifically 'Mastodon' which is a decentralised version of Twitter. To achieve this, a web collection tool is being developed to gather and analyse data from Mastodon. This project aims to perform a statistical analysis that characterises usage of Mastodon and design techniques to help Mastodon instances to identify malicious behaviour and resilience threats to the system. The output is a large-scale characterisation of user behaviour and resiliience threats to the infrastructure. This project is led by Dr Gareth Tyson, Dr Emiliano De Cristofaro and Professor Nishanth Sastry.