Institutions
|
About Us
|
Help
|
Gaeilge
0
1000
Home
Browse
Advanced Search
Search History
Marked List
Statistics
A
A
A
Show search options
Hide search options
Search using:
All
Any
None of these
Exact Phrase
in
Keyword (All Fields)
Title
Author
Subject
Institution
Funder
All
Any
None of these
Exact Phrase
in
Keyword (All Fields)
Title
Author
Subject
Institution
Funder
All
Any
None of these
Exact Phrase
in
Keyword (All Fields)
Title
Author
Subject
Institution
Funder
From
2104
2027
2021
2020
2019
2018
2017
2016
2015
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
1966
1965
1964
1963
1962
1961
1960
1959
1958
1957
1956
1955
1954
1953
1952
1951
1950
1949
1948
1947
1946
1945
1944
1943
1942
1941
1940
1939
1938
1937
1936
1935
1934
1933
1932
1931
1930
1929
1928
1927
1926
1925
1924
1923
1922
1921
1920
1919
1918
1917
1916
1915
1914
1913
1912
1911
1910
1909
1908
1907
1906
1905
1904
1903
1902
1901
1900
1899
1898
1897
1896
1895
1894
1893
1892
1891
1890
1889
1888
1887
1886
1885
1884
1883
1882
1881
1880
1879
1878
1877
1876
1875
1874
1873
1872
1871
1870
1869
1868
1867
1866
1865
1864
1863
1862
1861
1860
1859
1858
1857
1856
1855
1854
1853
1852
1851
1850
1849
1846
1842
1840
1839
1835
1827
1825
1821
1820
1818
1817
1815
1812
1811
1810
1809
1808
1807
1806
1805
1804
1803
1802
1801
1800
1792
1790
1770
1713
1111
1000
To
2104
2027
2021
2020
2019
2018
2017
2016
2015
2014
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
1966
1965
1964
1963
1962
1961
1960
1959
1958
1957
1956
1955
1954
1953
1952
1951
1950
1949
1948
1947
1946
1945
1944
1943
1942
1941
1940
1939
1938
1937
1936
1935
1934
1933
1932
1931
1930
1929
1928
1927
1926
1925
1924
1923
1922
1921
1920
1919
1918
1917
1916
1915
1914
1913
1912
1911
1910
1909
1908
1907
1906
1905
1904
1903
1902
1901
1900
1899
1898
1897
1896
1895
1894
1893
1892
1891
1890
1889
1888
1887
1886
1885
1884
1883
1882
1881
1880
1879
1878
1877
1876
1875
1874
1873
1872
1871
1870
1869
1868
1867
1866
1865
1864
1863
1862
1861
1860
1859
1858
1857
1856
1855
1854
1853
1852
1851
1850
1849
1846
1842
1840
1839
1835
1827
1825
1821
1820
1818
1817
1815
1812
1811
1810
1809
1808
1807
1806
1805
1804
1803
1802
1801
1800
1792
1790
1770
1713
1111
1000
Optionally, filter by:
(Leave unchecked to search all fields)
Item Type
Book
Book chapter
Conference item
Contribution to newspaper/magazine
Doctoral thesis
Journal article
Master thesis (research)
Master thesis (taught)
Multimedia
Patent
Report
Review
Working paper
Other
Peer Review Status
Peer reviewed
Non peer reviewed
Unknown
Institution
All Ireland Public Health Repository
Connacht-Ulster Alliance
Dublin City University
Dublin Institute of Technology
Dundalk Institute of Technology
Lenus
Marine Institute
Mary Immaculate College
Maynooth University
NUI Galway
Royal College of Surgeons in Ireland
Teagasc
Trinity College Dublin
University College Cork
University College Dublin
University of Limerick
Funder
Enterprise Ireland (EI)
Environmental Protection Agency (EPA)
Health Research Board (HRB)
Higher Education Authority (HEA)
Irish Aid
Irish Research Council for Humanities and Social Sciences (IRCHSS)
Irish Research Council for Science Engineering and Technology (IRCSET)
Marine Institute
Science Foundation Ireland (SFI)
Teagasc
Language
Irish
English
Bulgarian
Catalan; Valencian
Chinese
Czech
Danish
Dutch; Flemish
Estonian
French
German
Greek, Modern (1453-)
Croatian
Interlingue; Occidental
Italian
Japanese
Korean
Lithuanian
Norwegian
Polish
Portuguese
Romanian; Moldavian; Moldovan
Spanish; Castilian
Serbian
Turkish
Vietnamese
Current Search:
All of 'computer' and 'science' in all fields;
3089 items found
Sort by
Relevance
Title
Author
Item type
Date
Institution
Peer review status
Language
Order
Ascending
Descending
25
50
100
per page
1
2
3
4
5
6
7
8
9
10
Bibtex
CSV
EndNote
RefWorks
RIS
XML
Displaying Results 101 - 125 of 3089 on page 5 of 124
Marked
Mark
Discovering genome expression patterns with self-organizing neural networks
(2002)
Azuaje, Francisco
Discovering genome expression patterns with self-organizing neural networks
(2002)
Azuaje, Francisco
Abstract:
TCD-CS-2002-30
Self-organizing neural networks represent a family of useful clusteringbased classification methods in several application domains. One such technique is the Kohonen Self-Organizing Feature Map (SOM) (Kohonen, 2001), which has become one of the most successful approaches to analysing genomic expression data. This model is relatively easy to implement and evaluate, computationally inexpensive and scalable. In addition, it exhibits significant advantages in comparison to other options. For instance, unlike hierarchical clustering it facilitates an automatic detection and inspection of clusters. Unlike Bayesian-based clustering it does not require prior hypotheses or knowledge about the data under consideration. Compared to the k-means clustering algorithm, the SOM exemplifies a robust and structured classification process. - [Introduction]
http://hdl.handle.net/2262/2558
Marked
Mark
Efficient Prediction-Based Validation for Document Clustering
(2006)
Greene, Derek; Cunningham, Padraig
Efficient Prediction-Based Validation for Document Clustering
(2006)
Greene, Derek; Cunningham, Padraig
Abstract:
TCD-CS-2006-22
Recently, stability-based techniques have emerged as a very promising solution to the problem of cluster validation. An inherent drawback of these approaches is the computational cost of generating and assessing multiple clusterings of the data. In this paper we present an efficient prediction-based validation approach suitable for application to large, high-dimensional datasets such as text corpora. We use kernel clustering to isolate the validation procedure from the original data. Furthermore, we employ a prototype reduction strategy that allows us to work on a reduced kernel matrix, leading to significant computational savings. To ensure that this condensed representation accurately reflects the cluster structures in the data, we propose a density-biased selection strategy. This novel validation process is evaluated on a large number of real and artificial datasets, where it is shown to consistently produce good estimates for the optimal number of clusters.
http://hdl.handle.net/2262/13501
Marked
Mark
Evaluating Density Forecasting Models
(2006)
Carney, Michael; Cunningham, Padraig
Evaluating Density Forecasting Models
(2006)
Carney, Michael; Cunningham, Padraig
Abstract:
TCD-CS-2006-21
Density forecasting in regression is gaining popularity as real world applications demand an estimate of the level of uncertainty in predictions. In this paper we describe the two goals of density forecasting1 sharpness and calibration. We review the evaluation methods available to a density forecaster to assess each of these goals and we introduce a new evaluation method that allows modelers to compare and evaluate their models across both of these goals simultaneously and identify the optimal model.
http://hdl.handle.net/2262/13502
Marked
Mark
An Evaluation of the Usefulness of Explanation in a CBR System for Decision Support in Bronchiolitis Treatment
(2006)
Doyle, Donal; Cunningham, Padraig
An Evaluation of the Usefulness of Explanation in a CBR System for Decision Support in Bronchiolitis Treatment
(2006)
Doyle, Donal; Cunningham, Padraig
Abstract:
TCD-CS-2006-17
The research presented here explores the hypothesis that the deployment and acceptance of decision support systems in medicine will be enhanced if the basis for the recommendation produced by the system is apparent. We describe a decision support system for advising on patients suffering from bronchiolitis. This system supports its recommendations with precedent cases selected to support the recommendation along with justification text that highlights aspects of these cases relevant to the query case. It also presents an estimate of its confidence in the recommendation. The main contribution of this paper is an evaluation of this system in a clinical context. The evaluation shows that this type of explanation does enhance the usefulness of the system for practitioners.
http://hdl.handle.net/2262/13503
Marked
Mark
SAMPLE: An On-Demand Probabilistic Routing Protocol for Ad-hoc Networks
(2004)
Curran, Eoin; Dowling, Jim
SAMPLE: An On-Demand Probabilistic Routing Protocol for Ad-hoc Networks
(2004)
Curran, Eoin; Dowling, Jim
Abstract:
TCD-CS-2004-03
Existing on-demand ad hoc routing protocols assume an idealised wireless network in which all links in the network are either on or off and where all functioning links are equally good. Such a model interprets the fraction of packets that are dropped due to contention or interference as broken links, which can in turn lead to increased routing traffic and radio contention. As an alternative to the traditional hop-count metric, this paper presents a new metric for capturing the cost of a route based on a statisical model of network links. To investigage the impact of using this cost metric, we present a probabilistic routing protocol. SAMPLE, inspired by reinforcement learning techniques. Different scenario-based performance evaluations of the protocol in NS-2 are presented. In comparisons with AODV and DSR, SAMPLE exhibits improved performance in both lossy and congested wireless networks.
http://hdl.handle.net/2262/13281
Marked
Mark
A Review of Active Learning and Co-Training in Text Classification
(2005)
Davy, Michael
A Review of Active Learning and Co-Training in Text Classification
(2005)
Davy, Michael
Abstract:
TCD-CS-2005-64
http://hdl.handle.net/2262/13298
Marked
Mark
A Review of the Resource Management Task in ATM Networks
(1995)
Naughton, Shane
A Review of the Resource Management Task in ATM Networks
(1995)
Naughton, Shane
Abstract:
TCD-CS-95-06
[Introduction] An ATM network is a connection-oriented, packet-switched network with the ability to transport heterogeneous services irrespective of characteristics across the same underlying physical network. ATM networks were designed with the extremely high quality and high speeds of existing transmission systems in mind, and the potential future trend in the development of a multitude of services with diverse communication requirements, resulting in a exible and future-safe network. This paper presents a review of the resource management tasks that are necessary to provide network users with their requested performance criteria for the diverse and wide-ranging services they use, while at the same time ensuring there is a sufficiently high and balanced utilisation of the network resources. Section 2 provides an overview of the ATM network and the concepts of virtual connections and virtual paths. Section 3 discusses the need for resource management ofATM network...
http://hdl.handle.net/2262/12881
Marked
Mark
Aontas: The CaberNet Technical Abstracts Service
(1995)
Taylor, Paul
Aontas: The CaberNet Technical Abstracts Service
(1995)
Taylor, Paul
Abstract:
TCD-CS-95-22
CaberNet is the ESPRIT network of excellence in distributed systems consisting of several European research groups. CaberNet has industrial affiliates who receive regular information about the research activities of CaberNet members. Most CaberNet members produce technical reports which are of interest to the industrial affiliates and other researchers world-wide. This document describes the design and implementation of a unified technical report service. A contributing CaberNet site just has to make a bibliography available on a local machine. This bibliography is retrieved by a central site and any new or revised records are placed into a database. These records are processed by a professional library cataloguer who ensures that the information is relevant, complete and correct. The processed records may be searched over the world-wide-web and may be used to generate a summary of recent technical reports that is given to the industrial affiliates.
http://hdl.handle.net/2262/12862
Marked
Mark
A spatial programming model for real global smart space applications
(2006)
MEIER, RENE; CAHILL, VINNY
A spatial programming model for real global smart space applications
(2006)
MEIER, RENE; CAHILL, VINNY
Abstract:
Global smart spaces are intended to provide their inhabitants with context-aware access to pervasive services and information relevant to large geographical areas. Transportation is one obvious domain for such global smart spaces since applications can be built to exploit the variety of sensor-rich systems that have been deployed to support urban traffic control and highway management as well as within individual vehicles. This paper presents a spatial programming model designed to provide a standardised way to build contextaware global smart space applications using information that is distributed across independent (legacy, sensor-enabled, and embedded) systems by exploiting the overlapping spatial and temporal attributes of the information maintained by these systems. The spatial programming model is based on a topographical approach to modelling space that enables systems to independently define and use potentially overlapping spatial context in a consistent manner and in contra...
http://hdl.handle.net/2262/32587
Marked
Mark
Bayesian network based trust management
(2006)
CAHILL, VINNY
Bayesian network based trust management
(2006)
CAHILL, VINNY
Abstract:
Trust is an essential component for secure collaboration in uncertain environments. Trust management can be used to reason about future interactions between entities. In reputation-based trust management, an entity?s reputation is usually built on ratings from those who have had direct interactions with the entity. In this paper, we propose a Bayesian network based trust management model. In order to infer trust in di#11;erent aspects of an entity?s behavior, we use multidimensional application specific trust values and each dimension is evaluated using a single Bayesian network. This makes it easy both to extend the model to involve more dimensions of trust and to combine Bayesian networks to form an opinion about the overall trustworthiness of an entity. Each entity can evaluate his peers according to his own criteria. The dynamic characteristics of criteria and of peer behavior can be captured by updating Bayesian networks. Risk is explicitly combined with trust to help users mak...
http://hdl.handle.net/2262/32566
Marked
Mark
Proximity-Based Service Discovery in Mobile Ad Hoc Networks
(2005)
CLARKE, SIOBHAN; CAHILL, VINNY; MEIER, RENE
Proximity-Based Service Discovery in Mobile Ad Hoc Networks
(2005)
CLARKE, SIOBHAN; CAHILL, VINNY; MEIER, RENE
Abstract:
Existing approaches to service discovery have been developed primarily for environments with a fixed network backbone and typically rely on centralized components being accessible to potential service clients at any given time. The characteristic lack of a designated service infrastructure in combination with the highly dynamic nature of the underlying network topology renders such discovery mechanisms unsuitable for mobile ad hoc environments. This paper presents an approach to the discovery of ad hoc services that exploits the fact that the relevance of such services is often limited to specific geographical scopes. Service providers define the areas (proximities) in which their services are available. Clients register interest in specific services and are subsequently informed whenever they enter a proximity within which these services are available. Since ad hoc services can be stationary or may be moving with the location of their mobile providers our approach supports discover...
http://hdl.handle.net/2262/33126
Marked
Mark
A dynamic proxy based architecture to support distributed Java objects in a mobile environment
(2002)
CAHILL, VINNY; HAAHR, MADS
A dynamic proxy based architecture to support distributed Java objects in a mobile environment
(2002)
CAHILL, VINNY; HAAHR, MADS
Abstract:
The need to dynamically modify running applications arises in systems that must adapt to changes in their environment, in updating long-running systems that cannot be halted and restarted, and in monitoring and debugging systems without the need to recompile and restart them. Relatively few architectures have explored the meaning and possibilities of applying behavioural modifications to already running applications without static preparation of the application. The desirable characteristics of an architecture for dynamic modification include support for non-invasive association of new behaviour with the application, support for modular reusable components encapsulating the new behaviour and support for dynamic association (and de-association) of new behaviour with any class or object of the application. The Iguana/J architecture explores unanticipated dynamic modification, and demonstrates how these characteristics may be supported in an interpreted language without extending the l...
http://hdl.handle.net/2262/32507
Marked
Mark
An overview of the Tigger object-support operating system framework
(1996)
CAHILL, VINNY
An overview of the Tigger object-support operating system framework
(1996)
CAHILL, VINNY
Abstract:
This paper describes the motivations for and main features of Tigger -- a framework for the construction of a family of object-support operating systems that can be tailored for use in a variety of different application domains. An important goal of the design of Tigger is that instantiations of the framework should be able to support (a number of) different object models in order to allow a range of object-oriented languages for distributed or persistent programming to be supported without unnecessary duplication of effort. A further goal of the design is that instantiations of the framework should be able to support the same object model in different ways depending on the requirements of the applications to be supported by those instantiations. This paper describes the main features of the Tigger framework that allow these goals to be realised.
http://hdl.handle.net/2262/32408
Marked
Mark
On Adaptation in Analogy
(1995)
Keane, Mark T.
On Adaptation in Analogy
(1995)
Keane, Mark T.
Abstract:
TCD-CS-95-24
When people use analogies to solve problems they form an analogical mapping between two domains of knowledge. This mapping may support inferences by analogy that suggest a novel solution to a problem. Several factors have been proposed to be important in selecting this mapping, from among several alternative mappings: structural factors (systematicity and structural consistency) and pragmatic factors (the exploitation of higher-order planning categories). We suggest another set of factors play a role in selecting mappings: adaptability factors. Specifically, if a mapped solution can be adapted easily to a problem, then it will be preferred over an alternative mapping that is less adaptable. Two experiments are reported which test the effects of pragmatic and adaptation factors; using a novel technique in which the story analogue used has two alternative plans either of which can be used to solve an insight problem. In Experiment 1, these plans were varied in terms o...
http://hdl.handle.net/2262/12860
Marked
Mark
Knowledge Engineering Requirements in Derivational Analogy
(1993)
Cunningham, Padraig; Slattery, Sean; Finn, Donal
Knowledge Engineering Requirements in Derivational Analogy
(1993)
Cunningham, Padraig; Slattery, Sean; Finn, Donal
Abstract:
TCD-CS-93-21
A major advantage in using a case-based approach to developing knowledge-based systems is that it can be applied to problems where a strong domain theory may be difficult to determine. However the development of case-based reasoning (CBR) systems that set out to support a sophisticated case adaptation process does require a strong domain model. The Derivational Analogy (DA) approach to CBR is a case in point. In DA the case representation contains a trace of the reasoning process involved in producing the solution for that case. In the adaptation process this reasoning trace is reinstantiated in the context of the new target case; this requires a strong domain model and the encoding of problem solving knowledge. In this paper we analyse this issue using as an example a CBR system called CoBRA that assists with the modelling tasks in numerical simulation.
http://hdl.handle.net/2262/12921
Marked
Mark
Diversity versus Quality in Classification Ensembles based on Feature Selection
(2000)
Cunningham, Padraig; Carney, John G.
Diversity versus Quality in Classification Ensembles based on Feature Selection
(2000)
Cunningham, Padraig; Carney, John G.
Abstract:
TCD-CS-2000-02
Feature subset-selection has emerged as a useful technique for creating diversity in ensembles ? particularly in classification ensembles. In this paper we argue that this diversity needs to be monitored in the creation of the ensemble. We propose an entropy measure of the outputs of the ensemble members as a useful measure of the ensemble diversity. Further, we show that using the associated conditional entropy as a loss function (error measure) works well and the entropy in the ensemble predicts well the reduction in error due to the ensemble. These measures are evaluated on a medical prediction problem and are shown to predict the performance of the ensemble well. We also show that the entropy measure of diversity has the added advantage that it seems to model the change in diversity with the size of the ensemble.
http://hdl.handle.net/2262/13079
Marked
Mark
The Sam Loyd 15-Puzzle
(2001)
Hayes, Richard
The Sam Loyd 15-Puzzle
(2001)
Hayes, Richard
Abstract:
TCD-CS-2001-24
This report presents an approach to solve Sam Loyd?s famous 15-puzzle. The 15 or sliding puzzle is traditionally represented as a 4 * 4 board with tiles numbered from 1 to 15 arranged in numerical order from top left to bottom right of the board. The object of the puzzle is to rearrange the tiles in this order from any solvable scrambled starting position. A solvable board is one where the start configuration is an even permutation of the solved board. This report describes an algorithm that generates solutions to solvable boards not only for the 15 puzzle, but also for boards of different dimensions. It adapts a divide and conquer approach outlined in Ian Parberry?s paper (1997). Firstly, a brief discussion on the definition and background of the puzzle is offered. An in-depth explanation of what constitutes a solvable puzzle board is presented - an analysis of even permutations, which is fundamental to generating a solvable board, is given. From this, an algorit...
http://hdl.handle.net/2262/13099
Marked
Mark
Mobile Agents - Smart Messages
(1997)
Hurst, Leon; Cunningham, Padraig
Mobile Agents - Smart Messages
(1997)
Hurst, Leon; Cunningham, Padraig
Abstract:
TCD-CS-1997-05
Wireless communication with Mobile Computing devices is known to be problematic. It is very different in character from conventional communication over wired networks. Since many distributed applications make assumptions about network characteristics, they may not be used in a hostile mobile environment. We are proposing a new kind of messaging system which incorporates adaptive behaviour into the messages themselves. We call these `Smart Messages?, and implement them using Mobile Agents. The metaphor we use is of a message being delivered by a courier (Mobile Agent), on a potentially unresolved route. The `intelligence? is in the messages themselves rather than in the network. The approach taken expands on the self-routing capabilities of current Mobile Agent systems such as Aglets or Telescript. We aim to provide structured support for handling the particular problems associated with wireless communications. These include very limited, variable and asymmetric ba...
http://hdl.handle.net/2262/12981
Marked
Mark
Code Generation with ?Visual Modeller?
(1997)
Fillatre, Patrick
Code Generation with ?Visual Modeller?
(1997)
Fillatre, Patrick
Abstract:
TCD-CS-1997-04
?Visual Modeller? is a CASE tool that permits end-users to model business processes, using the associated method ?Visual Modelling?. It is also able to generate the database and a ?front-end? application from this model. Although this generation looks like an integrated process to the user, an intermediary step exists. From the model of the business process, ?Visual Modeller? first builds a model of the application that will be generated (ie a model of the future system), and then from this model, generates the database and the application. This document describes the code generation process in ?Visual Modeller?. How, from the business process model, it builds a model of the application, called the generic application. The generic application is greatly independent of the implementation environment; in fact, it is not built for any particular environment, but rather in a generic environment, which characteristics are predefined, independently of the characteristic...
http://hdl.handle.net/2262/12982
Marked
Mark
Producing Accurate Interpretable Clusters from High-Dimensional Data
(2005)
Greene, Derek; Padraig, Cunningham
Producing Accurate Interpretable Clusters from High-Dimensional Data
(2005)
Greene, Derek; Padraig, Cunningham
Abstract:
TCD-CS-2005-42
The primary goal of cluster analysis is to produce clusters that accurately reflect the natural groupings in the data. A second objective that is important for high-dimensional data is to identify features that are descriptive of the clusters. In addition to these requirements, we often wish to allow objects to be associated with more than one cluster. In this paper we present a technique, based on the spectral co-clustering model, that is effective in meeting these objectives. Our evaluation on a range of text clustering problems shows that the proposed method yields accuracy superior to that afforded by existing techniques, while producing cluster descriptions that are amenable to human interpretation.
http://hdl.handle.net/2262/13318
Marked
Mark
Practical Solutions to the Problem of Diagonal Dominance in Kernel Document Clustering
(2006)
Greene, Derek; Cunningham, Padraig
Practical Solutions to the Problem of Diagonal Dominance in Kernel Document Clustering
(2006)
Greene, Derek; Cunningham, Padraig
Abstract:
TCD-CS-2006-04
In supervised kernel methods, it has been observed that the performance of the SVM classifier is poor in cases where the diagonal entries of the Gram matrix are large relative to the off-diagonal entries. This problem, referred to as diagonal dominance, often occurs when certain kernel functions are applied to sparse high-dimensional data, such as text corpora. In this paper we investigate the implications of diagonal dominance for unsupervised kernel methods, specifically in the task of document clustering. We discuss a selection of strategies for addressing this issue, and evaluate their effectiveness in producing more accurate and stable clusterings.
http://hdl.handle.net/2262/13518
Marked
Mark
ECUE: A Spam Filter that Uses Machine Learning to Track Concept Drift
(2006)
Delany, Sarah Jane; Cunningham, Padraig
ECUE: A Spam Filter that Uses Machine Learning to Track Concept Drift
(2006)
Delany, Sarah Jane; Cunningham, Padraig
Abstract:
TCD-CS-2006-05
While text classification has been identified for some time as a promising application area for Artificial Intelligence, so far few deployed applications have been described. In this paper we present a spam filtering system that uses example-based machine learning techniques to train a classifier from examples of spam and legitimate email. This approach has the advantage that it can personalise to the specifics of the user?s filtering preferences. This classifier can also automatically adjust over time to account for the changing nature of spam (and indeed changes in the profile of legitimate email). A significant software engineering challenge in developing this system was to ensure that it could interoperate with existing email systems to allow easy managment of the training data over time. This system has been deployed and evaluated over an extended period and the results of this evaluation are presented here.
http://hdl.handle.net/2262/13506
Marked
Mark
Predicting Probability Distributions for Surf Height Using an Ensemble of Mixture Density Networks
(2006)
Carney, Michael; Cunningham, Padraig; Dowling, Jim
Predicting Probability Distributions for Surf Height Using an Ensemble of Mixture Density Networks
(2006)
Carney, Michael; Cunningham, Padraig; Dowling, Jim
Abstract:
TCD-CS-2006-06
There is a range of potential applications of Machine Learning where it would be more useful to predict the probability distribution for a variable rather than simply the most likely value for that variable. In meteorology and in finance it is often important to know the probability of a variable falling within (or outside) different ranges. In this paper we consider the prediction of surf height with the objective of predicting if it will fall within a given `surfable? range. Prediction problems such as this are considerably more difficult if the distribution of the phenomenon is significantly different from a normal distribution. This is the case with the surf data we have studied. To address this we use an ensemble of mixture density networks to predict the probability density function. Our evaluation shows that this is an effective solution. We also describe a web-based application that presents these predictions in a usable manner.
http://hdl.handle.net/2262/13505
Marked
Mark
Calibrating Probability Density Forecasts with Multi-objective Search
(2006)
Carney, Michael; Cunningham, Padraig
Calibrating Probability Density Forecasts with Multi-objective Search
(2006)
Carney, Michael; Cunningham, Padraig
Abstract:
TCD-CS-2006-07
In this paper, we show that the optimization of density forecasting models for regression in machine learning can be formulated as a multi-objective problem.We describe the two objectives of sharpness and calibration and suggest suitable scoring metrics for both.We use the popular negative log-likelihood as a measure of sharpness and the probability integral transform as a measure of calibration. We show how optimization on negative log-likelihood alone often results in sub-optimal models. To solve this problem we introduce a multi-objective evolutionary optimization framework that can produce better density forecasts from a prediction users perspective. Our experiments show improvements over state-of-the-art approaches.
http://hdl.handle.net/2262/13504
Marked
Mark
Adaptive Offset Subspace Self-Organizing Map: An Application to Handwritten Digit Recognition
(2006)
Zheng, Huicheng; Cunningham, Padraig; Tsymbal, Alexey
Adaptive Offset Subspace Self-Organizing Map: An Application to Handwritten Digit Recognition
(2006)
Zheng, Huicheng; Cunningham, Padraig; Tsymbal, Alexey
Abstract:
TCD-CS-2006-36
An Adaptive-Subspace Self-Organizing Map (ASSOM) can learn a set of ordered linear subspaces which correspond to invariant classes. However the basic ASSOMcannot properly learn linear manifolds that are shifted away from the origin of the input space. In this paper, we propose an improvement on ASSOM to amend this deficiency. The new network, named AOSSOM for Adaptive Offset Subspace Self-Organizing Map, minimizes a projection error function in a gradient-descent fashion. In each learning step, the winning module and its neighbors update their offset vectors and basis vectors of the target manifolds towards the negative gradient of the error function. We show by experiments that the AOSSOM can learn clusters aligned on linear manifolds shifted away from the origin and separate them accordingly. The proposed AOSSOM is applied to handwritten digit recognition and shows promising results.
http://hdl.handle.net/2262/13459
Displaying Results 101 - 125 of 3089 on page 5 of 124
1
2
3
4
5
6
7
8
9
10
Bibtex
CSV
EndNote
RefWorks
RIS
XML
Item Type
Book (3)
Book chapter (212)
Conference item (1165)
Contribution to newspaper/m... (1)
Doctoral thesis (217)
Journal article (745)
Master thesis (research) (52)
Master thesis (taught) (79)
Report (215)
Review (2)
Working paper (9)
Other (389)
Institution
Dublin City University (175)
NUI Galway (114)
Maynooth University (808)
Trinity College Dublin (859)
University College Cork (137)
University College Dublin (207)
University of Limerick (519)
Dublin Institute of Technology (128)
Lenus (2)
Dundalk Institute of Techno... (129)
Marine Institute (1)
Teagasc (2)
Royal College of Surgeons i... (5)
Connacht-Ulster Alliance (3)
Peer Review Status
Peer reviewed (1898)
Non peer reviewed (282)
Unknown (909)
Year
2021 (17)
2020 (87)
2019 (113)
2018 (91)
2017 (99)
2016 (102)
2015 (71)
2014 (131)
2013 (154)
2012 (231)
2011 (225)
2010 (233)
2009 (304)
2008 (150)
2007 (157)
2006 (160)
2005 (169)
2004 (126)
2003 (83)
2002 (69)
2001 (53)
2000 (43)
1999 (50)
1998 (24)
1997 (22)
built by Enovation Solutions