search
Include:
218 Research products, page 1 of 22

  • Publications
  • Research software
  • Other research products
  • 2012-2021
  • Open Access
  • Part of book or chapter of book
  • GB
  • FI
  • English
  • CORE (RIOXX-UK Aggregator)

10
arrow_drop_down
Relevance
arrow_drop_down
  • Publication . Part of book or chapter of book . Conference object . 2019
    Open Access English
    Authors: 
    Lara S. G. Piccolo; Somya Joshi; Evangelos Karapanos; Tracie Farrell;
    Country: Cyprus

    Part 12: Workshops; International audience; The manipulation of information and the dissemination of “fake news” are practices that trace back to the early records of human history. Significant changes in the technological environment enabling ubiquity, immediacy and considerable anonymity, have facilitated the spreading of misinformation in unforeseen ways, raising concerns around people’s (mis)perception of social issues worldwide. As a wicked problem, limiting the harm caused by misinformation goes beyond technical solutions, requiring also regulatory and behavioural changes. This workshop proposes to unpack the challenge at hand by bringing together diverse perspectives to the problem. Based on participatory design principles, it will challenge participants to critically reflect the limits of existing socio-technical approaches and co-create scenarios in which digital platforms support misinformation resilience.

  • Open Access English
    Authors: 
    Malgorzata A. Grzegorczyk; Pantea Lotfian; William J. Nuttall;
    Publisher: Springer

    In this chapter we explore the future for innovation in two related, but distinct, sectors. We consider the linkages between medical technology(MedTech) and agricultural technology (Agri-Tech) innovation in the UK. We ask and discuss questions: Who are the key actors in the innovation systems of Medtech and Agri-Tech in the UK? What are the core technologies driving the current waves of innovation in these two sectors? Can one industry learn from the other? Where is the scope for cooperation and synergies? We notice that both sectors are technologically linked through foundational technologies underpinning the majority of the observed innovation e.g. big data, AI, IoT and robotics. The outputs of these technologies rely crucially on digital data for insight and decision support. However, Agri-Tech benefits from less complex stakeholder issues regarding data security and privacy. Both sectors are important to the UK going forwards, and both will be exposed to Brexit and the consequences of the COVID pandemic. Our discussion on the future of innovation should be of particular interest to start-up leaders, entrepreneurs, investors, managers and policy-makers in MedTech, Agri-Tech and cognate sectors.

  • Publication . Part of book or chapter of book . 2015
    Open Access English
    Authors: 
    Kevin Swingler;
    Publisher: Springer

    The Multilayer Perceptron (MLP) is a neural network architecture that is widely used for regression, classification and time series forecasting. One often cited disadvantage of the MLP, however, is the difficulty associated with human understanding of a particular MLP’s function. This so called black box limitation is due to the fact that the weights of the network reveal little about structure of the function they implement. This paper proposes a method for understanding the structure of the function learned by MLPs that model functions of the class \(f:\{-1,1\}^n \rightarrow \mathbb {R}^m\). This includes regression and classification models. A Walsh decomposition of the function implemented by a trained MLP is performed and the coefficients analysed. The advantage of a Walsh decomposition is that it explicitly separates the contribution to the function made by each subset of input neurons. It also allows networks to be compared in terms of their structure and complexity. The method is demonstrated on some small toy functions and on the larger problem of the MNIST handwritten digit classification data set.

  • Open Access English
    Authors: 
    Sylvest, Matthew E.; Dixon, John C.; Conway, Susan J.; Patel, Manish R.; McElwaine, Jim N.; Hagermann, Axel; Barnes, Adam;
    Publisher: The Geological Society of London
    Project: EC | EPN2020-RI (654208), EC | UPWARDS (633127)

    Martian gullies were initially hypothesized to be carved by liquid water, due to their resemblance to gullies on Earth. Recent observations have highlighted significant sediment transport events occurring in Martian gullies at times and places where CO2 ice should be actively sublimating. Here we explore the role of CO2 sublimation in mobilizing sediment through laboratory simulation. In our previous experimental work, we reported the first observations of sediment slope movement triggered by the sublimation of CO2 frost. We used a Mars regolith simulant near the angle of repose. The current study extends our previous work by including two additional substrates, fine and coarse sand, and by testing slope angles down to 10°. We find that the Mars regolith simulant is active down to 17°, the fine sand is active only near the angle of repose and the coarse sand shows negligible movement. Using an analytical model, we show that under Martian gravity motion should be possible at even lower slope angles. We conclude that these mass-wasting processes could be involved in shaping Martian gullies at the present day and intriguingly the newly reported CO2-creep process could provide an alternative explanation for putative solifluction lobes on Mars.

  • Publication . Conference object . Part of book or chapter of book . 2016
    Open Access English
    Authors: 
    Trung Huynh; Yulan He; Allistair Willis; Stefan Rueger;
    Publisher: COLING
    Country: United Kingdom

    We study the problem of detecting sentences describing adverse drug reactions (ADRs) and frame the problem as binary classification. We investigate different neural network (NN) architectures for ADR classification. In particular, we propose two new neural network models, Convolutional Recurrent Neural Network (CRNN) by concatenating convolutional neural networks with recurrent neural networks, and Convolutional Neural Network with Attention (CNNA) by adding attention weights into convolutional neural networks. We evaluate various NN architectures on a Twitter dataset containing informal language and an Adverse Drug Effects (ADE) dataset constructed by sampling from MEDLINE case reports. Experimental results show that all the NN architectures outperform the traditional maximum entropy classifiers trained from n-grams with different weighting strategies considerably on both datasets. On the Twitter dataset, all the NN architectures perform similarly. But on the ADE dataset, CNN performs better than other more complex CNN variants. Nevertheless, CNNA allows the visualisation of attention weights of words when making classification decisions and hence is more appropriate for the extraction of word subsequences describing ADRs.

  • Publication . Part of book or chapter of book . 2019
    Open Access English
    Authors: 
    June Barrow-Green;
    Publisher: Springer

    This chapter is based on the talk that I gave in August 2018 at the ICM in Rio de Janeiro at the panel on "The Gender Gap in Mathematical and Natural Sciences from a Historical Perspective". It provides some examples of the challenges and prejudices faced by women mathematicians during last two hundred and fifty years. I make no claim for completeness but hope that the examples will help to shed light on some of the problems many women mathematicians still face today.

  • Publication . Part of book or chapter of book . Conference object . 2017
    Open Access English
    Authors: 
    Steven Arthur; Haijiang Li; Robert John Lark;
    Publisher: HAL CCSD

    Part 2: Production Information Systems; International audience; The current dominant computing mode in the AEC (Architecture, Engineering and Construction) domain is standalone based, causing fragmentation and fundamental interoperability problems. This makes the collaboration required to deal with the interconnected and complex tasks associated with a sustainable and resilient built environment extremely difficult.This article aims to discuss how the latest computing technologies can be leveraged for the AEC domain and Building Information Modelling (BIM) in particular. These technologies include Cloud Computing, the Internet of Things and Big Data Analytics.The data rich BIM domain will be analysed to identify relevant characteristics, opportunities and the likely challenges. A clear case will be established detailing why BIM needs these technologies and how they can be brought together to bring about a paradigm shift in the industry.Having identified the potential application of new technologies, a future platform will be proposed. It will carry out large scale, real-time processing of data from all stakeholders. The platform will facilitate the collaborative interpretation, manipulation and analysis of data for the whole lifecycle of building projects. It will be flexible, intelligent and able to autonomously execute analysis and choose the relevant tools. This will form a base for a step-change for computing tools in the AEC domain.

  • Publication . Part of book or chapter of book . 2014
    Open Access English
    Authors: 
    Miriam Fernandez; A. Elizabeth Cano; Harith Alani;
    Publisher: Springer International Publishing
    Project: EC | TRIVALENT (740934), EC | TRIVALENT (740934)

    Social Media is commonly used by policing organisations to spread the word on crime, weather, missing person, etc. In this work we aim to understand what attracts citizens to engage with social media policing content. To study these engagement dynamics we propose a combination of machine learning and semantic analysis techniques. Our initial research, performed over 3,200 posts from @dorsetpolice Twitter account, shows that writing longer posts, with positive sentiment, and sending them out before 4pm, was found to increase the probability of attracting attention. Additionally, posts about weather, roads and infrastructures, mentioning places, are also more likely to \ud attract attention.

  • Publication . Part of book or chapter of book . Conference object . 2016
    Open Access English
    Authors: 
    Faber, Wolfgang; Vallati, Mauro; Cerutti, Federico; Giacomin, Massimiliano;
    Country: United Kingdom

    Optimization—minimization or maximization—in the lattice of subsets is a frequent operation in Artificial Intelligence tasks. Examples are subset-minimal model-based diagnosis, nonmonotonic reasoning by means of circumscription, or preferred extensions in abstract argumentation. Finding the optimum among many admissible solutions is often harder than finding admissible solutions with respect to both computational complexity and methodology. This paper addresses the former issue by means of an effective method for finding subset-optimal solutions. It is based on the relationship between cardinality-optimal and subset-optimal solutions, and the fact that many logic-based declarative programming systems provide constructs for finding cardinality-optimal solutions, for example maximum satisfiability (MaxSAT) or weak constraints in Answer Set Programming (ASP). Clearly each cardinality-optimal solution is also a subset-optimal one, and if the language also allows for the addition of particular restricting constructs (both MaxSAT and ASP do) then all subset-optimal solutions can be found by an iterative computation of cardinality-optimal solutions. As a showcase, the computation of preferred extensions of abstract argumentation frameworks\ud using the proposed method is studied.

  • Publication . Part of book or chapter of book . Conference object . 2015
    Open Access English
    Authors: 
    Helen Caldwell; Neil Smith;
    Country: United Kingdom

    In September 2014, the computing curriculum in English schools changed to one with a much greater emphasis on computer science. However, 66% of existing ICT teachers are non-specialist and require significant continuing professional development (CPD) to deliver this new curriculum. One initiative to provide this is the Computing At School (CAS) Master Teacher programme. This paper describes some physical computing projects that were used in training a cohort of Master Teachers, preparing them to deliver both improved lessons in classrooms and CPD tailored for the requirements of their peers.

Send a message
How can we help?
We usually respond in a few hours.