Dublin University

Glasnevin, Ireland

Dublin University

Glasnevin, Ireland
Time filter
Source Type

Larrucea X.,University of the Basque Country | O'Connor R.V.,Dublin University | Colomo-Palacios R.,Ostfold University College | Laporte C.Y.,École de Technologie Supérieure of Montreal
IEEE Software | Year: 2016

ISO/IEC 29110 is becoming the set of software-process guidelines most widely adopted by very small entities.

Magdy W.,Dublin University | Darwish K.,Microsoft
Proceedings of the 2010 10th International Conference on Intelligent Systems Design and Applications, ISDA'10 | Year: 2010

Recent library digitization projects attempt to provide large collections of printed material from varying sources in a searchable format. The scanned documents are typically processed using Optical Character Recognition (OCR), which typically introduces errors in the text. This paper proposes a technique for correction of OCR degraded text that is independent of character-level OCR errors, and hence independent of scanned document source. It is based on language modeling in conjunction with a uniform character model that uses edit distance only. The technique compares well to state-of-the-art correction techniques that are based on language modeling and source-specific character error models. Although the proposed technique yielded lower correction effectiveness, its impact on retrieval effectiveness is statistically significant and at par with state-of-the-art correction techniques. The main requirement of the proposed technique is the training of a "good" language model matching genre, style, and temporal coverage. The advantage of being independent of character level errors is clear in applications were printed documents vary in source, font, and degradation level. © 2010 IEEE.

O'Connor R.,Dublin University | Hughes G.,Dublin University | Kauerauf T.,IMEC | Ragnarsson L.-A.,IMEC
IEEE International Reliability Physics Symposium Proceedings | Year: 2010

In this work we present the time dependent dielectric breakdown (TDDB) characteristics of LaO capped HfO2 layers with an equivalent oxide thickness of 8Å. The layers show maximum operating voltages in excess of 1V. Such high reliability can be attributed to very high Weibull slopes. We examine the origin of the high slopes by a detailed study of the evolution of the stress induced leakage current with time, temperature and stress voltage. © 2010 IEEE.

Vereshchagina E.,Dublin University | McGlade D.,Dublin University | Glynn M.,Dublin University | Ducree J.,Dublin University
Proceedings of the 16th International Conference on Miniaturized Systems for Chemistry and Life Sciences, MicroTAS 2012 | Year: 2012

A 3D hybrid microfluidic platform with integrated polycarbonate track-etched membranes has been developed. The platform enables cell culture and screening of chemotoxic reagents whilst the cell viability analysis has been carried out off-chip. Diffusive and convective modes of cross-membrane drug transfer have been established. In either mode, we successfully demonstrated on-chip incubation of HL60 myeloid leukemia cells and investigation after exposure to toxic mitomycin C (MMC) in the range of concentrations from 0 to 50 mM. The results of the viability assay on-a-chip were found to be consistent with a standard curve obtained off-chip and indicated ca. 40 % cell survival at MMC concentration of 50 mM.

Murphy E.,Dublin University | Bates E.,Dublin University | Fitzpatrick D.,Dublin University
ASSETS'10 - Proceedings of the 12th International ACM SIGACCESS Conference on Computers and Accessibility | Year: 2010

Visual mathematic notation provides a succinct and unambiguous description of the structure of mathematical formulae in a manner that is difficult to replicate through the linear channels of synthesized speech and Braille. It is proposed that the use of auditory cues can enhance accessibility to mathematical material and reduce common ambiguities encountered through spoken mathematics. However, the use of additional complex hierarchies of non-speech sounds to represent the structure and scope of equations may be cognitively demanding to process. This can detract from the users' understanding of the mathematical content. In this paper, a new system is presented, which uses a mixture of non-speech auditory cues, modified speech (spearcons) and binaural spatialization to disambiguate the structure of mathematical formulae. A design study, involving an online survey with 56 users, was undertaken to evaluate an existing set of auditory cues and to brainstorm alternative ideas and solutions from users before implementing modified designs and conducting a separate controlled evaluation. It is proposed that by involving a wide number of users in the creative design process, intuitive auditory cues will be implemented with the potential to enhance spoken mathematics for visually impaired users. © 2010 ACM.

Wilkins P.,Dublin University | Smeaton A.F.,Dublin University | Ferguson P.,Dublin University
SIGIR 2010 Proceedings - 33rd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval | Year: 2010

Content-Based Multimedia Information Retrieval (CBMIR) systems which leverage multiple retrieval experts (En) often employ a weighting scheme when combining expert results through data fusion. Typically however a query will comprise multiple query images (Im) leading to potentially N × M weights to be assigned. Because of the large number of potential weights, existing approaches impose a hierarchy for data fusion, such as uniformly combining query image results from a single retrieval expert into a single list and then weighting the results of each expert. In this paper we will demonstrate that this approach is sub-optimal and leads to the poor state of CBMIR performance in benchmarking evaluations. We utilize an optimization method known as Coordinate Ascent to discover the optimal set of weights (|E n| · |Im|) which demonstrates a dramatic difference between known results and the theoretical maximum. We find that imposing common combinatorial hierarchies for data fusion will half the optimal performance that can be achieved. By examining the optimal weight sets at the topic level, we observe that approximately 15% of the weights (from set |En| · |Im|) for any given query, are assigned 70%-82% of the total weight mass for that topic. Furthermore we discover that the ideal distribution of weights follows a log-normal distribution. We find that we can achieve up to 88% of the performance of fully optimized query using just these 15% of the weights. Our investigation was conducted on TRECVID evaluations 2003 to 2007 inclusive and ImageCLEFPhoto 2007, totalling 181 search topics optimized over a combined collection size of 661,213 images and 1,594 topic images. © 2010 ACM.

Ismail I.,Dublin University | Mazlan S.A.,University of Technology Malaysia | Olabi A.G.,Dublin University
Advanced Materials Research | Year: 2010

In this study, a testing rig in squeeze was designed and developed with the ability to conduct various tests especially for quasi-static squeezing at different values of magnetic field strength. Finite Element Method Magnetics (FEMM) was utilized to simulate the magnetic field distribution and magnetic flux lines generation from electromagnetic coil to the testing rig. Tests were conducted with two types of MR fluid. MRF-132DG was used to obtain the behaviour of MR fluid, while synthesized epoxy-based MR fluid was used for investigating the magnetic field distribution with regards to particle chains arrangement. Simulation results of the rig design showed that the magnetic flux density was well distributed across the tested materials. Magnetic flux lines were aligned with force direction to perform squeeze tests. Preliminary experimental results showed that stress-strain pattern of MR fluids were in agreement with previous results. The epoxy-based MR samples produced excellent metallographic samples for carbonyl iron particles distributions and particle chain structures investigation. © (2010) Trans Tech Publications.

Magdy W.,Dublin University | Jones G.J.F.,Dublin University
SIGIR 2010 Proceedings - 33rd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval | Year: 2010

Information retrieval (IR) evaluation scores are generally designed to measure the effectiveness with which relevant documents are identified and retrieved. Many scores have been proposed for this purpose over the years. These have primarily focused on aspects of precision and recall, and while these are often discussed with equal importance, in practice most attention has been given to precision focused metrics. Even for recall-oriented IR tasks of growing importance, such as patent retrieval, these precision based scores remain the primary evaluation measures. Our study examines different evaluation measures for a recall-oriented patent retrieval task and demonstrates the limitations of the current scores in comparing different IR systems for this task. We introduce PRES, a novel evaluation metric for this type of application taking account of recall and the user's search effort. The behaviour of PRES is demonstrated on 48 runs from the CLEF-IP 2009 patent retrieval track. A full analysis of the performance of PRES shows its suitability for measuring the retrieval effectiveness of systems from a recall focused perspective taking into account the user's expected search effort. © 2010 ACM.

Kennedy A.,Dundalk Institute of Technology | Wang X.,Dublin University
IEEE Transactions on Very Large Scale Integration (VLSI) Systems | Year: 2014

Packet classification is used by networking equipment to sort packets into flows by comparing their headers to a list of rules, with packets placed in the flow determined by the matched rule. A flow is used to decide a packet's priority and the manner in which it is processed. Packet classification is a difficult task due to the fact that all packets must be processed at wire speed and rulesets can contain tens of thousands of rules. The contribution of this paper is a hardware accelerator that can classify up to 433 million packets per second when using rulesets containing tens of thousands of rules with a peak power consumption of only 9.03 W when using a Stratix III field-programmable gate array (FPGA). The hardware accelerator uses a modified version of the HyperCuts packet classification algorithm, with a new pre-cutting process used to reduce the amount of memory needed to save the search structure for large rulesets so that it is small enough to fit in the on-chip memory of an FPGA. The modified algorithm also removes the need for floating point division to be performed when classifying a packet, allowing higher clock speeds and thus obtaining higher throughputs. © 1993-2012 IEEE.

Hurrell E.,Dublin University | Smeaton A.F.,Dublin University
ACM International Conference Proceeding Series | Year: 2013

Investigations into combining context and recommendation has resulted in much fruitful research which has improved recommender systems. Such contextual information has come in many forms and been used in different ways, successfully offering better in-situ suggestions. Factors such as location, time of recommendation, etc. have proven themselves as useful contributors to exploiting context. One issue, however, is the importance placed on each aspect of context, especially as new forms of recommendation are developed. Context is traditionally incorporated into recommenders at design-time, as a filter or as an integral part of how users are modelled, but the importance placed on each aspect is not often examined. Social recommenders and systems that draw on the wealth of data present in social networks frequently have access to far more contextual factors than traditional recommenders, making user relationships to these factors all the more important. The main contribution of this paper is to provide an examination of contextual priorities from the social web, which prove useful to recommender research in the area. This ontological examination of context shows that users have different priorities when it comes to context with a large variation in the suitability of each contextual factor in predicting good recommendations. In addition, this paper presents and discusses an approach to individually tailoring context ontologies (allowing for dynamically generated context sets), evaluating contextual factors in recommending from the social web. Copyright 2013 ACM.

Loading Dublin University collaborators
Loading Dublin University collaborators