Neil Hurley
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Neil Hurley has a total of 124 co-authors. Here are the top co-authors by number of shared publications:
- Aonghus Lawlor
University College Dublin, Dublin, Ireland
18 shared publications - Elias Tragos
University College Dublin
14 shared publications G.C.M. Silvestre
11 shared publications- Barry Smyth
University College Dublin, Dublin, Ireland
9 shared publications Mi Zhang
Sch. of Comput. Sci., Fudan Univ., Shanghai, China
9 shared publications- Diarmuid O'REilly Morgan
University College Dublin
8 shared publications
Latest Publications
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1170-1174
Transfer Learning via User-Item Graph Convolution for Enhanced Cross-Domain Recommendation
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1170-1174
Open AccessTransfer Learning via User-Item Graph Convolution for Enhanced Cross-Domain Recommendation
- Open Access
- 31 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Does Energy-efficiency Mean Carbon-awareness? A Critical Evaluation of Sustainable Serverless Edge Scheduling
- 31 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessDoes Energy-efficiency Mean Carbon-awareness? A Critical Evaluation of Sustainable Serverless Edge Scheduling
- Open Access
- 22 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 1302-1306
SlateLLM: Distilling LLM Semantics into Session-Aware Slate Recommendation without Inference Overhead
- 22 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 1302-1306
Open AccessSlateLLM: Distilling LLM Semantics into Session-Aware Slate Recommendation without Inference Overhead
- Open Access
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 4076-4080
NodeRec+: A Lightweight Framework for Federated Recommender Systems
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 4076-4080
Open AccessNodeRec+: A Lightweight Framework for Federated Recommender Systems
- Open Access
- 29 April 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Recommender Systems
Don't Get Bored: Enhancing Scalability and Diversity in Session-Based Slate Recommendation
Reinforcement learning (RL) has demonstrated great potential to improve slate-based recommender systems by optimizing long-term user engagement. However, addressing the combinatorial action space in slate recommendations remains challenging. Recent work decomposes slate Q-values into item-wise Q-values, improving the tractability of value-based methods to learn the model. But in scenarios with a large item pool and a resource-intensive value function like deep neural networks, the action selection process still incurs substantial computational costs. Slow training might be tolerable, but high costs during action selection could hinder real-time deployment. To address this issue, this paper introduces an actor method that reduces Q-function evaluations to a subset of items, significantly cutting inference time for practical deployment. The research suggests acquiring representations at both item and slate levels, strategically identifying a specific item subset for slate composition. The proposed methodologies are assessed over different simulated user engagement behaviors: users certain about preferences (“decisive” behavior) and those more exploratory or bored users, losing interest with repetitive content exposure (“explorative” behavior). Empirical evaluation shows that the proposed approach achieves comparable user engagement with a value-based policy across behaviors. Meanwhile, it notably enhances serving time while recommending diverse topic slates, thus demonstrating its potential effectiveness and efficiency in real-world applications.
- 29 April 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Recommender Systems
Open AccessDon't Get Bored: Enhancing Scalability and Diversity in Session-Based Slate Recommendation
Reinforcement learning (RL) has demonstrated great potential to improve slate-based recommender systems by optimizing long-term user engagement. However, addressing the combinatorial action space in slate recommendations remains challenging. Recent work decomposes slate Q-values into item-wise Q-values, improving the tractability of value-based methods to learn the model. But in scenarios with a large item pool and a resource-intensive value function like deep neural networks, the action selection process still incurs substantial computational costs. Slow training might be tolerable, but high costs during action selection could hinder real-time deployment. To address this issue, this paper introduces an actor method that reduces Q-function evaluations to a subset of items, significantly cutting inference time for practical deployment. The research suggests acquiring representations at both item and slate levels, strategically identifying a specific item subset for slate composition. The proposed methodologies are assessed over different simulated user engagement behaviors: users certain about preferences (“decisive” behavior) and those more exploratory or bored users, losing interest with repetitive content exposure (“explorative” behavior). Empirical evaluation shows that the proposed approach achieves comparable user engagement with a value-based policy across behaviors. Meanwhile, it notably enhances serving time while recommending diverse topic slates, thus demonstrating its potential effectiveness and efficiency in real-world applications.
- 4 April 2025
- Published by Springer Nature
- p. 246-254
DiffGR: A Discrete Diffusion-Based Model for Personalised Recommendation by Reconstructing User-Item Bipartite Graphs
- 4 April 2025
- Published by Springer Nature
- p. 246-254
DiffGR: A Discrete Diffusion-Based Model for Personalised Recommendation by Reconstructing User-Item Bipartite Graphs
- Open Access
- 26 November 2024
- Published by JMIR Publications Inc. in JMIR mHealth and uHealth
- Vol. 12 (1) , e49443
An Evaluation of the Effect of App-Based Exercise Prescription Using Reinforcement Learning on Satisfaction and Exercise Intensity: Randomized Crossover Trial
Background: The increasing prevalence of sedentary lifestyles has prompted the development of innovative public health interventions, such as smartphone apps that deliver personalized exercise programs. The widespread availability of mobile technologies (eg, smartphone apps and wearable activity trackers) provides a cost-effective, scalable way to remotely deliver personalized exercise programs to users. Using machine learning (ML), specifically reinforcement learning (RL), may enhance user engagement and effectiveness of these programs by tailoring them to individual preferences and needs. Objective: The primary aim was to investigate the impact of the Samsung-developed i80 BPM app, implementing ML for exercise prescription, on user satisfaction and exercise intensity among the general population. The secondary objective was to assess the effectiveness of ML-generated exercise programs for remote prescription of exercise to members of the public. Methods: Participants were randomized to complete 3 exercise sessions per week for 12 weeks using the i80 BPM mobile app, crossing over weekly between intervention and control conditions. The intervention condition involved individualizing exercise sessions using RL, based on user preferences such as exercise difficulty, selection, and intensity, whereas under the control condition, exercise sessions were not individualized. Exercise intensity (measured by the 10-item Borg scale) and user satisfaction (measured by the 8-item version of the Physical Activity Enjoyment Scale) were recorded after the session. Results: In total, 62 participants (27 male and 42 female participants; mean age 43, SD 13 years) completed 559 exercise sessions over 12 weeks (9 sessions per participant). Generalized estimating equations showed that participants were more likely to exercise at a higher intensity (intervention: mean intensity 5.82, 95% CI 5.59‐6.05 and control: mean intensity 5.19, 95% CI 4.97‐5.41) and report higher satisfaction (RL: mean satisfaction 4, 95% CI 3.9-4.1 and baseline: mean satisfaction 3.73, 95% CI 3.6-3.8) in the RL model condition. Conclusions: The findings suggest that RL can effectively increase both the intensity with which people exercise and their enjoyment of the sessions, highlighting the potential of ML to enhance remote exercise interventions. This study underscores the benefits of personalized exercise prescriptions in increasing adherence and satisfaction, which are crucial for the long-term effectiveness of fitness programs. Further research is warranted to explore the long-term impacts and potential scalability of RL-enhanced exercise apps in diverse populations. This study contributes to the understanding of digital health interventions in exercise science, suggesting that personalized, app-based exercise prescriptions may be more effective than traditional, nonpersonalized methods. The integration of RL into exercise apps could significantly impact public health, particularly in enhancing engagement and reducing the global burden of physical inactivity.
- 26 November 2024
- Published by JMIR Publications Inc. in JMIR mHealth and uHealth
- Vol. 12 (1) , e49443
Open AccessAn Evaluation of the Effect of App-Based Exercise Prescription Using Reinforcement Learning on Satisfaction and Exercise Intensity: Randomized Crossover Trial
Background: The increasing prevalence of sedentary lifestyles has prompted the development of innovative public health interventions, such as smartphone apps that deliver personalized exercise programs. The widespread availability of mobile technologies (eg, smartphone apps and wearable activity trackers) provides a cost-effective, scalable way to remotely deliver personalized exercise programs to users. Using machine learning (ML), specifically reinforcement learning (RL), may enhance user engagement and effectiveness of these programs by tailoring them to individual preferences and needs. Objective: The primary aim was to investigate the impact of the Samsung-developed i80 BPM app, implementing ML for exercise prescription, on user satisfaction and exercise intensity among the general population. The secondary objective was to assess the effectiveness of ML-generated exercise programs for remote prescription of exercise to members of the public. Methods: Participants were randomized to complete 3 exercise sessions per week for 12 weeks using the i80 BPM mobile app, crossing over weekly between intervention and control conditions. The intervention condition involved individualizing exercise sessions using RL, based on user preferences such as exercise difficulty, selection, and intensity, whereas under the control condition, exercise sessions were not individualized. Exercise intensity (measured by the 10-item Borg scale) and user satisfaction (measured by the 8-item version of the Physical Activity Enjoyment Scale) were recorded after the session. Results: In total, 62 participants (27 male and 42 female participants; mean age 43, SD 13 years) completed 559 exercise sessions over 12 weeks (9 sessions per participant). Generalized estimating equations showed that participants were more likely to exercise at a higher intensity (intervention: mean intensity 5.82, 95% CI 5.59‐6.05 and control: mean intensity 5.19, 95% CI 4.97‐5.41) and report higher satisfaction (RL: mean satisfaction 4, 95% CI 3.9-4.1 and baseline: mean satisfaction 3.73, 95% CI 3.6-3.8) in the RL model condition. Conclusions: The findings suggest that RL can effectively increase both the intensity with which people exercise and their enjoyment of the sessions, highlighting the potential of ML to enhance remote exercise interventions. This study underscores the benefits of personalized exercise prescriptions in increasing adherence and satisfaction, which are crucial for the long-term effectiveness of fitness programs. Further research is warranted to explore the long-term impacts and potential scalability of RL-enhanced exercise apps in diverse populations. This study contributes to the understanding of digital health interventions in exercise science, suggesting that personalized, app-based exercise prescriptions may be more effective than traditional, nonpersonalized methods. The integration of RL into exercise apps could significantly impact public health, particularly in enhancing engagement and reducing the global burden of physical inactivity.
- 14 November 2024
- Published by Springer Nature
- p. 58-67
Entity Matching with Large Language Models as Weak and Strong Labellers
- 14 November 2024
- Published by Springer Nature
- p. 58-67
Entity Matching with Large Language Models as Weak and Strong Labellers
- Open Access
- 21 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 3902-3906
RecPrompt: A Self-tuning Prompting Framework for News Recommendation Using Large Language Models
News recommendations heavily rely on Natural Language Processing (NLP) methods to analyze, understand, and categorize content, enabling personalized suggestions based on user interests and reading behaviors. Large Language Models (LLMs) like GPT-4 have shown promising performance in understanding natural language. However, the extent of their applicability to news recommendation systems remains to be validated. This paper introduces RecPrompt, the first self-tuning prompting framework for news recommendation, leveraging the capabilities of LLMs to perform complex news recommendation tasks. This framework incorporates a news recommender and a prompt optimizer that applies an iterative bootstrapping process to enhance recommendations through automatic prompt engineering. Extensive experimental results with 400 users show that RecPrompt can achieve an improvement of 3.36% in AUC, 10.49% in MRR, 9.64% in nDCG@5, and 6.20% in nDCG@10 compared to deep neural models. Additionally, we introduce TopicScore, a novel metric to assess explainability by evaluating LLM's ability to summarize topics of interest for users. The results show LLM's effectiveness in accurately identifying topics of interest and delivering comprehensive topic-based explanations.
- 21 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 3902-3906
Open AccessRecPrompt: A Self-tuning Prompting Framework for News Recommendation Using Large Language Models
News recommendations heavily rely on Natural Language Processing (NLP) methods to analyze, understand, and categorize content, enabling personalized suggestions based on user interests and reading behaviors. Large Language Models (LLMs) like GPT-4 have shown promising performance in understanding natural language. However, the extent of their applicability to news recommendation systems remains to be validated. This paper introduces RecPrompt, the first self-tuning prompting framework for news recommendation, leveraging the capabilities of LLMs to perform complex news recommendation tasks. This framework incorporates a news recommender and a prompt optimizer that applies an iterative bootstrapping process to enhance recommendations through automatic prompt engineering. Extensive experimental results with 400 users show that RecPrompt can achieve an improvement of 3.36% in AUC, 10.49% in MRR, 9.64% in nDCG@5, and 6.20% in nDCG@10 compared to deep neural models. Additionally, we introduce TopicScore, a novel metric to assess explainability by evaluating LLM's ability to summarize topics of interest for users. The results show LLM's effectiveness in accurately identifying topics of interest and delivering comprehensive topic-based explanations.
- Open Access
- 8 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1152-1157
Exploring Coresets for Efficient Training and Consistent Evaluation of Recommender Systems
Recommender systems have achieved remarkable success in various web applications, such as e-commerce, online advertising, and social media, harnessing the power of big data. To attain optimal model performance, recommender systems are typically trained on very large datasets, with substantial numbers of users and items. However, large datasets often present challenges in terms of processing time and computational resources. Coreset selection offers a method for obtaining a reduced yet representative subset from vast datasets, thereby enhancing the efficiency of training machine learning algorithms. Nevertheless, little research has been conducted to explore the practical implications of different coreset selection approaches on the performance of recommender systems algorithms. In this paper, we systematically investigate the impact of various coreset selection techniques. We evaluate the performance of the resulting coresets using inductive recommendation models which allow for consistent evaluations to be performed. The experimental results demonstrate that coreset methods are a powerful and useful approach for obtaining reduced datasets which preserve the properties of the large original dataset and have competitive performance compared to the time required to train with the full dataset.
- 8 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1152-1157
Open AccessExploring Coresets for Efficient Training and Consistent Evaluation of Recommender Systems
Recommender systems have achieved remarkable success in various web applications, such as e-commerce, online advertising, and social media, harnessing the power of big data. To attain optimal model performance, recommender systems are typically trained on very large datasets, with substantial numbers of users and items. However, large datasets often present challenges in terms of processing time and computational resources. Coreset selection offers a method for obtaining a reduced yet representative subset from vast datasets, thereby enhancing the efficiency of training machine learning algorithms. Nevertheless, little research has been conducted to explore the practical implications of different coreset selection approaches on the performance of recommender systems algorithms. In this paper, we systematically investigate the impact of various coreset selection techniques. We evaluate the performance of the resulting coresets using inductive recommendation models which allow for consistent evaluations to be performed. The experimental results demonstrate that coreset methods are a powerful and useful approach for obtaining reduced datasets which preserve the properties of the large original dataset and have competitive performance compared to the time required to train with the full dataset.