Derek Bridge

University College Cork

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Belén Díaz-Agudo

Universidad Complutense de Madrid

Giovanni Gabbolini

University College Cork

Nirmalie Wiratunga

Robert Gordon University

Mesut Kaya

Aalborg University

Anjana Wijekoon

Robert Gordon University

David Corsar

Robert Gordon University

Kyle Martin

Robert Gordon University

Anne Liret

Orange (France)

Juan A. Recio-García

Universidad Complutense de Madrid

Francesco Ricci⋆

Free University of Bozen-Bolzano

Cooperative Institutions

University College Cork

University College Dublin

Robert Gordon University

Universidad Complutense de Madrid

University of York

Free University of Bozen-Bolzano

Usibelli Coal Mine (United States)

Dublin City University

Trinity College Dublin

Software (Spain)

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Denoising Dictionary Learning Against Adversarial Perturbations

arXiv (Cornell University) (2018)

John Mitro Derek Bridge Steven Prestwich

We propose denoising dictionary learning (DDL), a simple yet effective technique as a protection measure against adversarial perturbations. We examined denoising dictionary learning on MNIST and CIFAR10 perturbed under two different perturbation techniques, fast gradient sign (FGSM) and jacobian saliency maps (JSMA). We evaluated it against five different deep neural networks (DNN) representing the building blocks of most recent architectures indicating a successive progression of model complexity of each other. We show that each model tends to capture different representations based on their architecture. For each model we recorded its accuracy both on the perturbed test data previously misclassified with high confidence and on the denoised one after the reconstruction using dictionary learning. The reconstruction quality of each data point is assessed by means of PSNR (Peak Signal to Noise Ratio) and Structure Similarity Index (SSI). We show that after applying (DDL) the reconstruction of the original data point from a noisy

MNIST database

Similarity (geometry)

Dictionary Learning

Source

Cite

Citations (3)

The Blogoduct System: A Just-In-Time Information Retrieval Assistant for Bloggers

Lecture notes in computer science (2010)

Ang Gao Derek Bridge

10.1007/978-3-642-17080-5_32

Cite

Citations (2)

Conversational Query Revision with a Finite User Profiles Model.

IIR eBooks (2012)

Henry Blanco Francesco Ricci⋆Derek Bridge

Information Recommendation is a conversational approach aimed at suggesting to the user how to reformulate his queries to a product catalogue in order to find the products that maximize his utility. In previous work, it was shown that, by observing the queries selected by the user among those suggested, the system can make inferences on the true user utility function and eliminate from the set of suggested queries those retrieving products with an inferior utility (dominated queries). The computation of the dominated queries was based on the solution of several linear programming problems, which represented a major computational bottleneck for the efficiency of the proposed solution. In this paper we propose a new technique for the computation of the dominated queries. It relies on the assumption that the set of possible user utility functions is finite. We show that under this assumption the computation of the query suggestions is simplified and the number of query suggestions is strongly reduced.

Lexicographical order

Result set

Source

Cite

Citations (4)

Recommending Personalized Query Revisions

Conference on Recommender Systems (2012)

Henry Blanco Francesco Ricci⋆Derek Bridge

Observing the queries selected by a user, among those suggested by a recommender system, one can infer constraints on the user’s utility function, and can avoid suggesting queries that retrieve products with an inferior utility, i.e., dominated queries. In this paper we propose a new e ffi cient technique for the computation of dominated queries. It relies on the system’s assumption that the number of possible profiles (or utility functions), of the users it may interact with, is finite. We show that making query suggestions is simplified, and the number of suggestions is strongly reduced. We also found that even if the system is not contemplating the true user profile, among the above mentioned finite set of profiles, its performance is still very close to the optimal one.

Result set

Source

Cite

Citations (4)

Automatic Playlist Continuation using Subprofile-Aware Diversification

Mesut Kaya Derek Bridge

The ACM RecSys Challenge 2018 involves the task of automatic playlist continuation (APC), aiming to help users to create and extend their own music playlists. In this paper, we explain teamrozik's approach to the Challenge. Our approach to APC is twofold: Cold-Start-APC for short playlists and SPAD-APC for other playlists. Cold-Start-APC is a rudimentary popularity-based recommender. SPAD-APC treats playlists as if they were user profiles. It builds an implicit matrix factorization model to generate initial recommendations. But it re-ranks those recommendations using SubProfile-Aware Diversification (SPAD), which is a personalized intent-aware diversification method. The SPAD re-ranking method aims to ensure that the final set of recommendations covers different interests or tastes in the playlists of the users, which we refer to as subprofiles. We show that such subprofiles do exist within playlists and we show that the SPAD method achieves higher precision than matrix factorization alone.

Popularity

Continuation

10.1145/3267471.3267472

Cite

Citations (8)

Report on the Twenty‐Second International Conference on Case‐Based Reasoning

AI Magazine (2015)

Derek Bridge Luc Lamontagne Enric Plaza

The Twenty‐Second International Conference on Case‐Based Reasoning was held from September 29 to October 1, 2014. This report summarizes the conference highlights.

10.1609/aimag.v36i2.2587

Cite

Citations (0)

Item-Based Explanations for User-Based Recommendations

Marius Kaminskas Fred Durao Derek Bridge

Source

Cite

Citations (0)

Retrieval through explanation: an abductive inference approach to relevance feedback

Ian Ruthven Mounia Lalmas C. J. van Rijsbergen Derek Bridge Ruth M. J. Byrne

Relevance feedback techniques are designed to automatically improve a system's representation of a query by using documents the user has marked as relevant. However, traditional relevance feedback models suffer from a number of limitations that restrict their potential in supporting information seeking. One of the major limitations of relevance feedback is that it does not incorporate behavioural aspects of information seeking - how and why users assess relevance. We propose that relevance feedback should be viewed as a process of explanation and demonstrate how this limitation of relevance feedback techniques can be overcome by a theory of relevance feedback based on abductive inference.

Relevance

Relevance Feedback

Abductive reasoning

Cite

Citations (4)

Accurate and Diverse Recommendations Using Item-Based SubProfiles.

The Florida AI Research Society (2018)

Mesut Kaya Derek Bridge

Source

Cite

Citations (6)

We're all going on a summer holiday: an exercise in non-cardinal case base retrieval