Assoc. Prof. David Lillis

Academics

Assoc. Prof. David Lillis

University College Dublin

www.lill.is

David Lillis is an Associate Professor in the School of Computer Science at University College Dublin (UCD). He has published peer-reviewed research papers in digital forensics, information retrieval, agent-oriented software engineering, component-based systems, and wireless sensor networks.

He is affiliated with the Beijing-Dublin International College and is a collaborator in the CeADAR centre for applied data analytics.

Research Output

Publications

2020
First-page preview of Assessing the Influencing Factors on the Accuracy of Underage Facial Age Estimation

Assessing the Influencing Factors on the Accuracy of Underage Facial Age Estimation

Felix Anda; Brett Becker; David Lillis; Nhien-An Le-Khac; Mark Scanlon

The 6th IEEE International Conference on Cyber Security and Protection of Digital Services (Cyber Security)

This study evaluates the influencing factors on the accuracy of underage facial age estimation using two cloud services, Microsoft Azure's Face API and Amazon Web Service's Rekognition service. The analysis of the VisAGe dataset reveals correlations between facial attributes and age estimation errors, identifying the most significant factors to be addressed in future age estimation modeling.

2019
First-page preview of Improving Borderline Adulthood Facial Age Estimation through Ensemble Learning

Improving Borderline Adulthood Facial Age Estimation through Ensemble Learning

Felix Anda; David Lillis; Aikaterini Kanta; Brett Becker; Elias Bou-Harb; Nhien-An Le-Khac; Mark Scanlon

The 8th International Workshop on Cyber Crime (IWCC), held at the 14th International Conference on Availability, Reliability and Security (ARES)

This paper presents an ensemble learning approach to improve facial age estimation for borderline adulthood cases. The authors develop a deep learning model (DS13K) and fine-tune it on the Deep Expectation (DEX) model to achieve an accuracy of 68% for the age group 16-17 years old, outperforming DEX by 4 times. The study also evaluates existing cloud-based facial age prediction services.

2019
First-page preview of Improving the Accuracy of Automated Facial Age Estimation to Aid CSEM Investigations

Improving the Accuracy of Automated Facial Age Estimation to Aid CSEM Investigations

Felix Anda; David Lillis; Aikaterini Kanta; Brett A. Becker; Elias Bou-Harb; Nhien-An Le-Khac; M. Scanlon

Digital Investigation Vol. 28 pp. S142

This study evaluates existing age prediction services and introduces a deep learning model, DS13K, to improve the accuracy of underage facial age estimation in child sexual exploitation material (CSEM) investigations. The model outperforms existing services, particularly in the borderline adulthood age range (16-17 years old), with an accuracy rate of 68%.

2018
First-page preview of Evaluating Automated Facial Age Estimation Techniques for Digital Forensics

Evaluating Automated Facial Age Estimation Techniques for Digital Forensics

Felix Anda; David Lillis; Nhien-An Le-Khac; Mark Scanlon

12th International Workshop on Systematic Approaches to Digital Forensics Engineering (SADFE), IEEE Security & Privacy Workshops

This paper evaluates existing automated facial age estimation techniques for digital forensics, highlighting their limitations and proposing a dataset generator to overcome the lack of sufficient sample images in specific age ranges. The study assesses the performance of offline and cloud-based models, releasing a tool to generate uniformly distributed random images by age and gender.

2018
First-page preview of Hierarchical Bloom Filter Trees for Approximate Matching

Hierarchical Bloom Filter Trees for Approximate Matching

David Lillis; Frank Breitinger; Mark Scanlon

Journal of Digital Forensics, Security and Law Vol. 13 pp. 81-96

This paper proposes the use of Hierarchical Bloom Filter Trees (HBFTs) to improve the runtime efficiency of approximate matching techniques in digital forensics. HBFTs reduce the number of pairwise comparisons required, achieving substantial speed gains while maintaining effectiveness. The authors evaluate the effectiveness of HBFTs using the MRSH-v2 algorithm and explore the effects of different configurations of HBFTs.

2018
First-page preview of Expediting MRSH-v2 Approximate Matching with Hierarchical Bloom Filter Trees

Expediting MRSH-v2 Approximate Matching with Hierarchical Bloom Filter Trees

David Lillis; Frank Breitinger; Mark Scanlon

Digital Forensics and Cyber Crime. ICDF2C 2017 Vol. 216 pp. 144-157

This paper presents an improvement to the MRSH-v2 approximate matching algorithm using Hierarchical Bloom Filter Trees (HBFT) to expedite the search process for digital forensic investigators. Experiments demonstrate substantial speed gains over the original MRSH-v2 while maintaining effectiveness.

2017
First-page preview of EviPlant: An Efficient Digital Forensic Challenge Creation, Manipulation, and Distribution Solution

EviPlant: An Efficient Digital Forensic Challenge Creation, Manipulation, and Distribution Solution

Mark Scanlon; Xiaoyu Du; David Lillis

Digital Investigation Vol. 20S pp. 29-36

EviPlant is a system designed to efficiently create, manipulate, store, and distribute digital forensic challenges for education and training. It allows educators to create evidence packages that can be integrated with base images, reducing the need for large, full-image files and making it easier to distribute challenges to students.

2016
First-page preview of Current Challenges and Future Research Areas for Digital Forensic Investigation

Current Challenges and Future Research Areas for Digital Forensic Investigation

David Lillis; Brett Becker; Tadhg O'Sullivan; Mark Scanlon

The 11th ADFSL Conference on Digital Forensics, Security and Law (CDFSL 2016) pp. 9-20

This paper explores the current challenges in digital forensic investigations, including the digital evidence backlog, and outlines future research areas to improve the process. The authors discuss the increasing complexity, diversity, and volume of digital evidence, as well as the need for standardization and automation in digital forensic tools and processes.

2016
First-page preview of On the Benefits of Information Retrieval and Information Extraction Techniques Applied to Digital Forensics

On the Benefits of Information Retrieval and Information Extraction Techniques Applied to Digital Forensics

David Lillis; Mark Scanlon

Advanced Multimedia and Ubiquitous Engineering: FutureTech & MUE pp. 641-647

This paper explores the application of Information Retrieval (IR) and Information Extraction (IE) techniques to digital forensics, highlighting their potential to improve the efficiency and effectiveness of investigations. The authors discuss the benefits of cloud-based digital forensic investigation platforms and the importance of precision and recall in different stages of an investigation.