ERA :: Home

ERA is a digital repository of original research produced at The University of Edinburgh. The archive contains documents written by, or affiliated with, academic authors, or units, based at Edinburgh that have sufficient quality to be collected and preserved by the Library, but which are not controlled by commercial publishers. Holdings include full-text digital doctoral theses, masters dissertations, project reports, briefing papers and out-of-print materials.

Information on current research activity including staff, projects and publications is available via the Edinburgh Research Explorer.

Now showing 1 - 40 of 41

Recent Submissions

Factors affecting the consumer’s intention to adopt Fintech in Islamic banking in Malaysia
(The University of Edinburgh. Management School and Economics. Economics, 2026-05-21) Binti Khairol Nizam, Aini Nur Hajjar; Ansell, Jonathan; Harrison, Tina; Yule, Jennifer; Universiti Utara Malaysia; Ministry of Higher Education, Malaysia
Open Banking leverages consumer data to enhance access to and management of financial services. However, its success depends heavily on consumers’ willingness to share data in among the banks and their third-party provider. This study explores the factors influencing consumer acceptance of Open Banking in the context of Islamic banking institutions in Malaysia, where religious and cultural values may shape perceptions of risk and trust differently. The major findings show that performance expectancy, perceived risk, and trust are significant predictors of consumer acceptance. Performance expectancy which is the consumers' belief in the usefulness of Open Banking was the strongest driver of acceptance. In contrast, perceived risk negatively impacted acceptance, while trust played a key role in encouraging data-sharing behaviour. A mixed-methods approach was adopted, starting with interviews with industry experts, followed by regression analysis of survey data from 118 respondents, and further validated using Structural Equation Modelling (SEM) via AMOS with a sample of 341 respondents. The data was purposefully collected on Malaysian Islamic banks to ensure that the differing approaches to risk and trust within the Muslim community were appropriately understood. The study concludes that consumer acceptance of Open Banking is shaped by both the perceived benefits of the service and the confidence that data will be managed securely and ethically. Building trust and addressing risk concerns are therefore essential to optimising data use and enhancing consumer participation in Open Banking within Islamic financial systems.
Unified framework for decomposing neural representations and analyzing specialization in language models
(The University of Edinburgh, 2026-05-21) Zhao, Zheng; Cohen, Shay; Webber, Bonnie; UKRI CDT in Natural Language Processing
The rise of large, pre-trained Transformer models has transformed Natural Language Processing (NLP), yet the internal mechanisms by which these models handle diverse and heterogeneous data remain insufficiently understood. This thesis addresses this gap by developing and applying a unified analytical framework to examine how such models represent, differentiate, and specialize for distinct subpopulations of data. The central contribution is the Model-Oriented Sub-population and Spectral Analysis (MOSSA) framework, which systematically contrasts a generalist model, trained on multiple domains, languages, or tasks, with a suite of specialist control models trained on individual subpopulations. Through a set of advanced matrix analysis techniques, MOSSA quantifies representational similarities layer by layer, revealing where and how knowledge encoding and adaptation occur within the model architecture. The framework is applied across three major studies of increasing complexity. The first investigates domain learning using Singular Vector Canonical Correlation Analysis (SVCCA) to assess how model capacity and data scale affect the encoding of domain-specific information. The findings show that larger models not only generalize across domains but also embed domain-specialist behavior within their internal representations, particularly for domain-specific vocabulary. The second study extends this approach to multilingual modeling. A joint matrix factorization method is introduced to analyze representational structures across 33 languages. The analysis uncovers systematic variation in the encoding of morphosyntactic information across layers, shaped by linguistic properties such as script and morphological complexity. Moreover, the learned representations align with cross-lingual task performance and yield linguistically meaningful phylogenetic structures. The third study explores the dynamics of massively multi-task instruction tuning in Large Language Models (LLMs). Using Centered Kernel Alignment (CKA) within MOSSA, we examine how an LLM represents over 60 NLP tasks. The results reveal a distinct architectural segmentation: early shared layers encode general-purpose features, intermediate transition layers rapidly acquire task-specific information, and later refinement layers optimize representations for precise task execution. Together, these studies establish a principled methodology for probing and interpreting the internal organization of large neural models. The thesis demonstrates that generalist language models systematically partition their representational space, forming specialized subspaces tailored to different data regimes. This work identifies where such specialization arises within model depth and clarifies the mechanisms underlying adaptation, multilinguality, and multi-task learning in contemporary NLP systems.
Balancing investment in clean heat and energy efficiency in Scottish housing retrofit
(BRE, 2026-05) Weeks, Caroline; Bruce-Konuah , Adorkor; Sinclair, Colin
This study aimed to ascertain whether there is consensus on an optimal balance between fabric efficiency and clean heat solutions for Scottish homes. The balance must deliver concurrently on the cross-sector agendas of net zero, fuel poverty, and public health.
Explain, interpret, and convince: Zhanran and his commentary Zhiguan fuxing chuanhong jue
(The University of Edinburgh. College of Humanities and Social Science. School of Divinity, 2026-05-21) Jin, Rusha; Appleton; Gentz, Joachim
This study is organised around the overarching question of why Jingxi Zhanran (711–782) is called the ‘Tiantai Master of Commentary’. Focusing on one of Zhanran’s key works, the Zhiguan fuxing chuanhongjue (hereafter ZFCJ), a commentary on one of the Tiantai School’s foundational texts, the Mohe zhiguan (hereafter MHZG). The study examines his commentarial enterprise and how exactly it contributed to the Tiantai school. The central inquiry of this study is: What does it mean to be the ‘Tiantai Master of Commentary’? The inquiry rests on two key premises: Zhanran’s mastery of Buddhist exegesis and his interpretations contributed to the continuity and development of the Tiantai tradition. To examine this, the study explores four sub-questions: 1.What made Zhanran a commentator, and why does the ZFCJ matter within his corpus? 2.What are Zhanran’s commentarial approach and methods? 3.How can Zhanran’s exegesis support the Tiantai tradition? 4.What does Zhanran’s case tell us about the Chinese Buddhist commentarial tradition? The four chapters of this study address the questions outlined above. Chapter One contextualises the creation of the ZFCJ, situating it among Zhanran’s other works and establishing its aim to preserve the correct understanding of the MHZG. It argues that while the ZFCJ may not be entirely original, it reflects a deep engagement with and transmission of Tiantai teachings. Chapters Two and Three analyse Zhanran’s commentarial techniques, showing his thorough explication of the MHZG and his systematic approach to interpreting Zhiyi’s exegesis. These chapters highlight Zhanran’s ability to reinforce Tiantai authority by skilfully integrating external sources and demonstrating his deep doctrinal knowledge. Chapter Four examines how Zhanran builds upon and extends Zhiyi’s commentarial methods by looking at his systematisation of Zhiyi’s exegetical principles in the MHZG, revealing his understanding of the hermeneutical process that bridges ‘text’ and ‘meaning’. Building on this analysis, the chapter explores the nature of commentarial methods and considers how the creative dimensions embedded within them should be understood. This study not only contributes to a more comprehensive understanding of Zhanran and the ZFCJ, and the intellectual history of Tiantai tradition in Chinese Buddhism, but also, through this case of a successful commentator who addressed complex issues with skilful exegesis, enriches our appreciation of how an individual could actively participate in and influence broader intellectual communities through commentarial creation. Furthermore, it offers a paradigm and raises noteworthy issues for future studies of Chinese Buddhist commentarial literature, which are meaningful to commentarial studies but also to cultural studies and religious studies more broadly.
Legal transplant of Anglo-American whistleblowing to Chile
(The University of Edinburgh. College of Humamities and Social Sciences, 2026-05-21) Martínez Rivera, Jorge; Cabrelli, David; Macgregor, Laura; Agencia Nacional de Investigación y Desarrollo (ANID), Chile
This Thesis examines the challenges and limits of the legal transplant of Anglo-American whistleblowing protections, particularly anti-retaliation, into the Chilean legal system. It argues that whistleblowing, as a legal category, is an original innovation of Anglo-American law that lacks a true functional equivalent in civil law jurisdictions, such Chile. Drawing on comparative analysis, the research demonstrates that whistleblowing protection in the United States is embedded in its employment discrimination framework and supported by procedural mechanisms, such as discovery and public interest conflicts in the workplace, that are absent in Chile. Through a combination of doctrinal analysis and comparative methodology, the Thesis first defines the concept of whistleblowing and traces its historical and linguistic origins in English law. It then assesses the Chilean legal framework’s capacity to receive and replicate Anglo-American whistleblowing rules, identifying significant mismatches in both substance and procedure. The study shows that Chile’s constitutional and statutory system of employment discrimination, which lacks a clear distinction between public interest disclosures and workplace grievances, is unsuited to hosting a functional equivalent of whistleblower protection. The findings suggest that transplanting whistleblowing protection into Chilean law requires more than adopting statutory language–it demands deeper institutional and cultural adjustments to achieve the intended function. From a comparative law methodology, this Thesis contributes to scholarship by illustrating the limitations of functionalist assumptions and the importance of historical and systemic context in the adoption of whistleblower protection in Chile.