Irish Universities
Not a member yet
78452 research outputs found
Sort by
Active Learning for Multi-Label Image Annotation
No full textActive learning is useful in situations where labeled data is scarce, unlabeled data is available and labeling has some cost associated with it. In such situations active learning helps by identifying a minimal set of items to label that will allow the training of an effective classifier. Thus active learning is appropriate for annotation tasks in multimedia, particularly in image labeling. In this paper we address the challenge of using active learning for multi-labeling of images in personal image collections. Multi-label learning covers situations where objects can have more than one class label and a learner is trained to assign multiple labels simultaneously. In this paper we report results on a learning system for labeling personal image collections that is both active and multi-label. The focus of the research has been to reduce the overall number of images that are presented to the user for labeling
An Analysis of Current Trends in CBR Research Using Multi-View Clustering
No full textThe European Conference on Case-Based Reasoning (CBR) in 2008 marked 15 years of international and European CBR conferences where almost seven hundred research papers were published. In this report we review the research themes covered in these papers and identify the topics that are active at the moment. The main mechanism for this analysis is a clustering of the research papers based on both co-citation links and text similarity. It is interesting to note that the core set of papers has attracted citations from almost three thousand papers outside the conference collection so it is clear that the CBR conferences are a sub-part of a much larger whole. It is remarkable that the research themes revealed by this analysis do not map directly to the sub-topics of CBR that might appear in a textbook. Instead they reflect the applications-oriented focus of CBR research, and cover the promising application areas and research challenges that are faced.Science Foundation Irelan
Decoding the kinematic and ionisation structure of axisymmetric nebulae
No full textThe aim of the thesis was to develop a code and pipeline for generating 3D photoionisation models of axisymmetric nebulae to better understand their ob- served structure and composition. This was achieved through a new code that amalgamated the features of popular (but separate) codes in astronomy; a 3D morpho-kinematic modelling application called Shape, a well known, large-scale spectral synthesis code called Cloudy and PyCloudy, a Python library that handles Cloudy. The resulting code presented in this thesis was called Py- Cross, an acronym for “PyCloudy Rendering of Shape Software”.
The steep learning curves traditionally experienced when developing and using new codes overshadows the necessity for a formal software development lifecycle. As software development and coding is becoming an essential skill for new astronomers, who are often required to create their own codes for specific purposes, employing and adhering to a software development lifecycle during developing will help meet milestones while managing complex projects, thus ensuring reliability and quality. Here, a formal software development lifecycle and test driven development approach is described and employed in the development of the PyCross code. A detailed account of the code development, installation, functionality, user interface and operational overview is given using both theoretical and actual stellar objects.
Creating photoionisation models for known planetary nebulae and novae can only be accomplished when there is sufficient information to determine the most significant physical parameters of the nebula and the central star. Presented here are novel approaches for various scientific methods/pipelines that can be employed with PyCross to generate photoionisation models. Over the course of this thesis PyCross has been used to develop, for the first time, 3D photoionisation models of novae V5668 Sagittarii (2015), V4362 Sagittarii (PTB 42) as well as planetary nebulae LoTr 1 and MyCn 18
Moral judgment as categorization (MJAC)
No full textObserved variability and complexity of judgments of “right” and “wrong” cannot be readily accounted for within extant approaches to understanding moral judgment. In response to this challenge, we present a novel perspective on categorization in moral judgment. Moral judgment as categorization (MJAC) incorporates principles of category formation research while addressing key challenges of existing approaches to moral judgment. People develop skills in making context-relevant categorizations. They learn that various objects (events, behaviors, people, etc.) can be categorized as morally right or wrong. Repetition and rehearsal result in reliable, habitualized categorizations. According to this skill-formation account of moral categorization, the learning and the habitualization of the forming of moral categories occur within goal-directed activity that is sensitive to various contextual influences. By allowing for the complexity of moral judgments, MJAC offers greater explanatory power than existing approaches while also providing opportunities for a diverse range of new research questions.Publishedpeer-reviewe
Policy Incentives as Behavioural Drivers of Beef Enterprises in Ireland: Where are the Kinks?
No full textThe current structure of agricultural production is still influenced by historical coupled
payments, even though it has been eight years since decoupled payments were introduced.
Much of the expansion in the Irish cattle herd that occurred during the era of the MacSharry
reforms is still visible. In this paper we consider the incentives associated with the Common
Agricultural Policy (CAP) over time in relation to production. Our primary focus is on
subsidies that were available to the beef sector, and we investigate the behavioural pressures
associated with these incentives. We have developed a Hypothetical microsimulation model
using a typical farm, based on plausible values taken from the Teagasc National Farm Survey
(NFS) 1995. We are investigating if subsidies available to the beef sector in Ireland through
the CAP since 1984 resulted in non-linearity in the Direct Payment Schedule faced by cattle
farmers, and if so where were these kinks and what were the behavioural pressures associated
with these incentives? Identifying non-linearity in the Direct Payment Schedule indicates
where incentives occurred. Large kinks are associated with large incentives at that point. We
calculated a total payment for each subsidy from 1984 to 2014, and constructed a Direct
Payment Schedule that varies by stocking rate. We find that subsidies, and in particular the
CAP reform payments of the MacSharry era introduced large discontinuities or kink points in
the Direct Payment Schedule of beef farmers, indicating that there were large incentives for
farmers to produce at or just before these points
Milk production per cow and per hectare of spring-calving dairy cows grazing swards differing in Lolium perenne L. ploidy and Trifolium repens L. composition
No full textGrazed grass is the cheapest feed available for dairy
cows in temperate regions; thus, to maximize profits,
dairy farmers must optimize the use of this high-quality
feed. Previous research has defined the benefits of
including white clover (Trifolium repens L.) in grass
swards for milk production, usually at reduced nitrogen
usage and stocking rate. The aim of this study was to
quantify the responses in milk production of dairy cows
grazing tetraploid or diploid perennial ryegrass (Lolium
perenne L.; PRG) sown with and without white clover
but without reducing stocking rate or nitrogen usage.
We compared 4 grazing treatments in this study: tetraploid
PRG-only swards, diploid PRG-only swards,
tetraploid with white clover swards, and diploid with
white clover swards. Thirty cows were assigned to each
treatment, and swards were rotationally grazed at a
farm-level stocking rate of 2.75 cows/ha and a nitrogen
fertilizer rate of 250 kg/ha annually. Sward white clover
content was 23.6 and 22.6% for tetraploid with white
clover swards and diploid with white clover swards, respectively.
Milk production did not differ between the
2 ploidies during this 4-yr study, but cows grazing the
PRG-white clover treatments had significantly greater
milk yields (+596 kg/cow per year) and milk solid
yields (+48 kg/cow per year) compared with cows grazing
the PRG-only treatments. The PRG-white clover
swards also produced 1,205 kg of DM/ha per year more
herbage, which was available for conserving and buffer
feeding in spring when these swards were less productive
than PRG-only swards. Although white clover is
generally combined with reduced nitrogen fertilizer use,
this study provides evidence that including white clover
in either tetraploid or diploid PRG swards, combined
with high levels of nitrogen fertilizer, can effectively
increase milk production per cow and per hectar
The Literature-Enactment-Process: Exploring narratives through performative conventions
No full textThis project promotes reading literature for students through a new approach termed the Literature-Enactment-Process (LEP) where students can gain access to and comprehend narratives and associated topics of inquiry through a range of phases, with drama-based conventions as a pivotal point. As a pedagogical tool, these performative strategies are embedded in a larger approach that combines individual and collaborative comprehension processes. The LEP seeks to explore literature interactively, in that the student’s individual views, the perceptions of others, and the text details are equally taken into account. Teaching literature should not remain restricted to correctly answering interpretative questions. If teachers demand only one “right” interpretation, learners are deprived of the enrichment and multiple meanings texts can generate. Students must be motivated to think and learn for themselves and for a world which is constantly changing, often to the detriment of our natural environment. For this purpose, the Literature and Ecology (LITECO) workshop was designed to fuse the study of literature and ecological learning using and exemplifying the LEP. At the University of Graz, the Literature-Enactment-Process was tested with current and future teachers as well as language arts students and positively evaluated as an interdisciplinary teaching approach for the (foreign) language classroom in secondary education
Process drama in the classroom: A case study of developing participation for advanced EAL learners in an international school
No full textThis paper reports on a study of the use of process drama in an international primary school in the Netherlands. The research investigated the extent to which using process drama could develop participation for advanced EAL learners. In addition, we sought to understand pupils’ perspectives. Using a qualitative methodology, we undertook a case study approach focusing on six advanced EAL learner pupils (9-10-year-olds). We implemented the process drama approach during a series of nine science lessons. We collated and analysed Video recording of lessons, the class teacher’s written observations, a research journal, two interviews and a focus group with the case study participants using an arts-based framework of participation, previously employed by Pérez-Moreno (2018). We deployed embodied research methods. The findings suggest that using process drama as a teaching methodology increased participation, but not immediately. In addition, pupils who had not previously spoken out in lessons began to volunteer their ideas. All case study pupils reported that they considered that their participation increased
Covid-19, non-Covid-19 and excess mortality rates not comparable across countries
No full textEvidence that more people in some countries and fewer in others are dying because of the pandemic, than is reflected by reported coronavirus disease 2019 (Covid-19) mortality rates, is derived from mortality data. Using publicly available databases, deaths attributed to Covid-19 in 2020 and all deaths for the years 2015–2020 were tabulated for 35 countries together with economic, health, demographic and government response stringency index variables. Residual mortality rates (RMR) in 2020 were calculated as excess mortality minus reported mortality rates due to Covid-19 where excess deaths were observed deaths in 2020 minus the average for 2015–2019. Differences in RMR are differences not attributed to reported Covid-19. For about half the countries, RMR\u27s were negative and for half, positive. The absolute rates in some countries were double those in others. In a regression analysis, population density and proportion of female smokers were positively associated with both Covid-19 and excess mortality while the human development index and proportion of male smokers were negatively associated with both. RMR was not associated with any of the investigated variables. The results show that published data on mortality from Covid-19 cannot be directly comparable across countries. This may be due to differences in Covid-19 death reporting and in addition, the unprecedented public health measures implemented to control the pandemic may have produced either increased or reduced excess deaths due to other diseases. Further data on cause-specific mortality is required to determine the extent to which residual mortality represents non-Covid-19 deaths and to explain differences between countries
Convergence of blockchain, autonomous agents, and knowledge graph to share electronic health records
No full textIn this article, we discuss a data sharing and knowledge integration framework through autonomous agents with blockchain for implementing Electronic Health Records (EHR). This will enable us to augment existing blockchain-based EHR Systems. We discuss how major concerns in the health industry, i.e., trust, security and scalability, can be addressed by transitioning from existing models to convergence of the three technologies blockchain, agent-based modeling, and knowledge graph in a decentralized ecosystem. Each autonomous agent is responsible for instantiating key processes, such as user authentication and authorization, smart contracts, and knowledge graph generation through data integration among the participating stakeholders in the network. We discuss a layered approach for the design of the proposed system leading to an enhanced, safer clinical decision-making system. This can pave the way toward more informed and engaged patients and citizens by delivering personalized healthcare