Digital Commons Network^™

Essays On Institutions And Human Capital, Xiaofan Zhu

Dissertations and Theses Collection (Open Access)

This dissertation investigates the critical role of political and culturally embedded institutions in shaping human capital investment and firm behavior, focusing on three distinct but related areas. Chapter 1 investigates how institutional quality influences educational choices in China, with a focus on behavioral responses to the anti-corruption campaign. The study finds that politically connected students from regions more a!ected by the campaign are significantly more likely to pursue further education – especially those with non-STEM majors, weaker academic performance, and stronger political ties. The analysis of underlying mechanisms suggests that as the employment environment became more equitable, connected individuals responded …

The Effects Of Income On Infant Health: Evidence From The Coal Boom And Bust, Chu Gek Chew

Dissertations and Theses Collection (Open Access)

An individual’s health at birth is an important component in human capital accumulation, posing significant implications both in the short- and long-run. I investigate how infant health outcomes were impacted during a period of successive positive and negative economic shocks led by the coal industry between the 1970s to 1990s. Using exogenous variation within counties’ pre-existing coal reserves, I instrument for the changes in economic opportunities during the coal boom and bust. I find some limited evidence that an increase in income improves infant health outcomes. The incidence of low birthweight is decreasing in earnings growth, and effects are stronger …

Does Moving From One-Story Building To Multi-Building Affect The Health Status Of The Elderly: Evidence From China’S Par Program, Lu Cao

Dissertations and Theses Collection (Open Access)

This study investigates the causal impact of relocating from single-story to multistory buildings on the health outcomes of elderly residents in China. Using China’s Poverty Alleviation Relocation (PAR) program as an exogenous policy shock, we employ instrumental variable and difference-in-differences approaches to establish causality. Our analysis of data from the China Health and Retirement Longitudinal Study (2011-2018) reveals a nuanced relationship between vertical living and elderly health. While relocation to multi-story buildings is associated with deterioration in self-reported physical health status, it significantly improves mental health conditions and life satisfaction among elderly movers. We identify two key mechanisms driving these …

Intraday Volatility In Financial Markets: Evidence From High-Frequency Data, Haolin Wang

Dissertations and Theses Collection (Open Access)

In this study, we conduct an analysis of intraday spot volatility using high-frequency data from the SPDR S&P 500 ETF (SPY). We begin with the assumption that the intraday volatility of asset prices exhibits time-variation. To capture this dynamic behavior, we construct proxies for the unobserved spot volatility by applying appropriate estimators to the high-frequency price data. Following the estimation procedure, we employ various forecasting models to generate volatility forecasts. Finally, we evaluate and compare the predictive performance of these models using established forecast evaluation metrics, and analyze the results.

Determinants Of Trading Prices In State-Owned Asset And Equity Transactions: An Empirical Analysis From China, Lixiao Chen

Dissertations and Theses Collection (Open Access)

This thesis examines the determinants of pricing outcomes in state-owned asset transactions using a comprehensive dataset of 3,638 transactions from a major transitional economy. Contrary to conventional expectations, we find that fixed assets consistently receive less favorable pricing compared to equity assets, with significantly lower odds of trading above professional valuation and higher odds of trading below valuation. This pattern persists across multiple specifications and remains robust to propensity score matching approaches that address selection concerns. I further demonstrate that state-owned buyers secure significant advantages in transaction pricing, with dramatically lower odds of paying premiums over valuation or listing prices …

Optimizing Fulfillment Operations For Online Retailing, Yujie Sun

Dissertations and Theses Collection (Open Access)

This thesis investigates order fulfillment operations for online retailing. Order fulfillment encompasses the entire process of receiving, processing, and delivering customer orders. The rapid growth of e-commerce has heightened the demand for efficient fulfillment operations. This thesis explores strategies for optimizing order fulfillment in three scenarios. The first scenario studies cellular bucket brigade, a method for coordinating workers in an assembly line or order-picking environment to improve workflow efficiency. The second scenario investigates zone picking, serial bucket brigades and cellular bucket brigades with hand-off times. The third scenario focuses on autonomous robotic systems within package sortation centers.

Chapter 2 investigates …

Nature Of Smartphone Use: The Underlying Mechanisms Between Nature Exposure And Urge To Use Social Media, Jia Yi Aw

Dissertations and Theses Collection (Open Access)

The urge to use social media is prevalent and costly in modern society. However, little has been researched on how nature exposure can manage urge to use social media to combat its negative consequences. It was hypothesized that nature exposure through Instagram posts in Study 1 and nature park in Study 2 reduces urge to use social media. The underlying mechanisms were tested in a theory-driven serial mediation model, in which positive emotions, attention restoration and self-regulation served as serial mediators. Study 1 employed Instagram experimental manipulation in a laboratory setting by randomly assigning 206 undergraduates to view Instagram accounts …

Essays On Financial Econometrics, Yuexuan Ren

Dissertations and Theses Collection (Open Access)

This dissertation consists of two papers that contribute to the theory of estimation, inference, and prediction using high-frequency financial data. In the first chapter, time-varying betas from return regressions, commonly used to measure systematic financial market risk, are studied using high-frequency data. A novel econometric framework is introduced for the nonparametric estimation of these betas. The optimal finite-sample inference, in a well-defined sense, is enabled by the local Gaussian property of a generic continuous-time benchmark model under the fixed-k framework. Furthermore, a uniform functional inference theory for spot regressions is developed. The proposed methodology is broadly applicable in empirical settings, …

Towards Reliable Ml: Data Attribution And Adversarial Robustness, Xiaosen Zheng

Dissertations and Theses Collection (Open Access)

Modern machine learning (ML) models achieve remarkable success, but face critical reliability challenges. This thesis advances two pillars of reliable ML systems: interpretability through data attribution and robustness against adversarial threats.

In the first part, we develop novel data attribution methods to elucidate the data-model relationship. We establish the critical role of memorization in model generalization through token-level influence analysis, extend sample-level attribution to diffusion models with effective approximation techniques, and introduce REGMIX, a group-level approach that predicts data mixture performance using small-scale experiments. These contributions provide practitioners with scalable tools to audit training data impacts across modalities.

The second …

Environmental Influences Of Zero-Sum Thinking And Its Outcome, Edison Sora Tan

Dissertations and Theses Collection (Open Access)

Zero-sum thinking refers to a generalized mentality or a belief that one person’s gain can only come at the expense of another person’s loss. In this research, I argue that people’s zero-sum thinking is, at least in part, influenced by the interacting characteristics of the environment, namely, resource scarcity and population density. This is because resource scarcity highlights people’s resource insecurity, particularly when population density accentuates potential competition for finite resources. Consequently, zero-sum thinking reduces trust and cooperation as people reduce their commitment to potentially risky social investments. Analysis with data from a large-scale field study and ecological data provided …

More Than Meets The Eye: Misaligned Expectations In Social Interactions For Individuals With Chronic Illness, Janice Patricia

Dissertations and Theses Collection (Open Access)

Despite the importance of social relationships for well-being, individuals with chronic illness often face barriers that deter them from initiating meaningful connections, such as perceived stigma, fear of rejection, or strained relationships upon a disclosure of their illness. These challenges reveal two key misalignments: overestimated negative (or underestimated positive) expectations of social interactions held by individuals with chronic illness, as well as perceptions held by potential acquaintances. The present study examined how these misalignments hinder social interactions and relationship initiation. Study 1 involved individuals with chronic illness and revealed that participants overestimated the likelihood of awkwardness, disrespect, and rejection, while …

Disentangling User Preferences Towards Self-Interpretable Recommender Systems, Nhu Thuat Tran

Dissertations and Theses Collection (Open Access)

Understanding user preferences remains a central challenge in recommender systems due to their inherently complex, unstructured, and multi-faceted nature, exacerbated by the sparsity of user interaction data. Traditional approaches often compress user interests into a single latent vector, overlooking the fact that user preferences are typically shaped by multiple underlying factors that differ across individuals. These latent drivers are not directly observable and must be discovered through unsupervised modeling, further complicated by limited historical interactions per user.

This dissertation addresses these challenges by introducing a principled framework for multiinterest modeling, which disentangles user behaviors into multiple latent factors to better …

What Makes An Experience Meaningful: A Process/Outcome Approach, Clement Yong Hao Lau

Dissertations and Theses Collection (Open Access)

Scant research has investigated how meaning judgement of an experience are derived; even fewer has examined how these judgements shift over time. To address these gaps, the present research introduced Process/Outcome Meaning Model (POMM)—which proposes that meaning can be derived from two distinct aspects of an experience: its process (i.e., the thoughts, feelings, and behaviors that unfold during the experience) and its outcome (i.e., the results or consequences that follow). POMM further suggests that changes in perceived outcomes may lead to corresponding shifts in meaning judgements over time. Across three studies, participants wrote about and reflected on their personal (positive …

Tailoring Transformer-Based Deep Learning For Code Generation And Translation, Imam Nur Bani Yusuf

Dissertations and Theses Collection (Open Access)

Software is increasingly pervasive in modern society, making the effective translation of human intent into code essential. Novice programmers often struggle with domain-specific code due to limited background knowledge, while experienced developers face challenges in maintaining evolving largescale codebases. Traditional pattern-based approaches address these issues, but such approaches are task-specific and require significant adaptation for different tasks. Transformer-based models offer a more flexible alternative, as the same architecture can be tailored for diverse programming tasks.

This dissertation investigates how Transformer-based models can be customized for various code generation and translation tasks. First, it introduces Transformer-based approaches that assist end-users with …

Formation And Impact Of Informal Interest Groups: The Role Of Clan Culture, Bin Wang

Dissertations and Theses Collection (Open Access)

This study explores the role of clan culture in the formation and dynamics of informal interest groups, particularly focusing on how multiple identities, such as clan identity, educational identity, and professional identity, interact to influence the intention to join the informal interest group, and subsequently influence individual outcomes along with group support. Through in-depth interviews with 21 members of the “Clan X Alumni Group”, an informal group with a same surname, this study investigates how clan culture fosters a strong sense of group identity and belonging, enhancing collaboration, knowledge sharing, and psychological well-being among members. It also extends social identity …

Empirical Program Evaluation In Operations Management, Kai Tian

Dissertations and Theses Collection (Open Access)

This dissertation focuses on advancing and applying empirical program evaluation methodologies in Operations Management (OM), defining it not as mere prediction but as rigorously establishing causal effects of interventions. Chapter 2 lays the theoretical groundwork, defining empirical program evaluation within OM, detailing its alignment with causal inference, and outlining both classical econometric techniques and emerging Causal Machine Learning (CML) approaches for mitigating endogeneity. Building on this foundation, Chapter 3 presents an empirical application, evaluating the causal impact of a customer targeting strategy using live-streaming advertisements, employing Propensity Score Matching (PSM) in conjunction with linear regression models to derive actionable insights …

Exploring The Relationship Between Intrinsic Capacity And Social Participation In Healthy Ageing: Evidence From Singapore, Li Ling Cheong, William Tov, Robin Wai Munn Choo, Micah Tan, Lay Khoon Lau, Wee Shiong Lim, Yew Yoong Ding, Paulin Tay Straughan

Research Collection School of Social Sciences

Objective: The World Health Organization has identified intrinsic capacity and functional ability as key constructs in enabling healthy ageing. However, functional ability is distinct from actual levels of social and physical participation, which research has shown to be associated with health and wellbeing. This study distinguishes between functional ability and actual levels of social and physical participation, and utilizes the International Classification of Functioning, Disability and Health framework to examine the relationship between IC and the two constructs. Methods: This study utilizes cross-sectional self-reported data from a nationally representative panel of Singapore residents aged 56–75 (N = 6,434). We performed …

Multiobjective Linear Ensembles For Robust And Sparse Training Of Few-Bit Neural Networks, Ambrogio Maria Bernardelli, Stefano Gualandi, Simone Milanesi, Hoong Chuin Lau, Neil Yorke-Smith

Research Collection School Of Computing and Information Systems

Training neural networks (NNs) using combinatorial optimization solvers has gained attention in recent years. In low-data settings, the use of state-of-the-art mixed integer linear programming solvers, for instance, has the potential to exactly train an NN while avoiding computing-intensive training and hyperparameter tuning and simultaneously training and sparsifying the network. We study the case of few-bit discrete-valued neural networks, both binarized neural networks (BNNs) whose values are restricted to ±1 and integer-valued neural networks (INNs) whose values lie in the range {−P,…,P}. Few-bit NNs receive increasing recognition because of their lightweight architecture and ability to run on low-power devices: for …

Enriching Automatic Test Case Generation By Extracting Relevant Test Inputs From Bug Reports, Wendkuuni C. Ouedraogo, Laura Plein, Kader Kabore, Andrew Habib, Jacques Klein, David Lo, Tegawende F. Bissyande

Research Collection School Of Computing and Information Systems

The quality of software is closely tied to the effectiveness of the tests it undergoes. Manual test writing, though crucial for bug detection, is time-consuming, which has driven significant research into automated test case generation. However, current methods often struggle to generate relevant inputs, limiting the effectiveness of the tests produced. To address this, we introduce BRMiner, a novel approach that leverages Large Language Models (LLMs) in combination with traditional techniques to extract relevant inputs from bug reports, thereby enhancing automated test generation tools. In this study, we evaluate BRMiner using the Defects4J benchmark and test generation tools such as …

Relation Prediction In Knowledge Graphs: A Self-Organizing Neural Network Approach, Budhitama Subagdja, Shanthoshigaa D, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Knowledge graphs (KGs) in specialized domains frequently suffer from incomplete information. While current relation prediction methods for KG completion typically rely on neural network-based representation learning, we present KG2ART---a novel self-organizing neural network that employs a fundamentally different approach. Instead of learning distributed representations, KG2ART encodes relation triples of knowledge graphs explicitly and performs parallel inference over the graph structure through bidirectional interactions between bottom-up activations and top-down pattern matching. Our comprehensive evaluation across five diverse KGs (Nations, UMLS, Kinship, CoDEx-M, and a jet engine technical KG) demonstrates that KG2ART consistently outperforms state-of-the-art baselines (TuckER, ComplEX, RESCAL, ConvE, CompGCN) in …

Enhancing Deliberativeness: Evaluating The Impact Of Multimodal Reflection Nudges, Shun Yi Yeo, Zhuoqun Jiang, Anthony Tang, Simon Tangi Perrault

Research Collection School Of Computing and Information Systems

Nudging participants with text-based reflective nudges enhances deliberation quality on online deliberation platforms. The effectiveness of multimodal reflective nudges, however, remains largely unexplored. Given the multi-sensory nature of human perception, incorporating diverse modalities into self-reflection mechanisms has the potential to better support various reflective styles. This paper explores how presenting reflective nudges of different types (direct: persona and indirect: storytelling) in different modalities (text, image, video and audio) affects deliberation quality. We conducted two user studies with 20 and 200 participants respectively. The first study identifies the preferred modality for each type of reflective nudges, revealing that text is most …

Creating Talking Points For Client Advisers At Banks To Promote Sustainable Investing, Wewe Zi Yi, Pradeep Varakantham, Alan Megargel

Research Collection School Of Computing and Information Systems

Environmental, social and governance (ESG) factors have become key nonfinancial factors for investors to evaluate companies with respect to understanding material risks and growth opportunities. While not mandatory, companies are providing ESG reports that outline progress in different ESG metrics (six broad metrics and 15 specific ones). Client advisers (CAs) read these reports to identify key metrics of interest to investors. Given the number of companies and investment products, however, it is not feasible for CAs to read all the reports, which can sometimes run into tens or hundreds of pages). The authors have developed multiple frameworks building on leading …

Prompting An Embodied Ai Agent: How Embodiment And Multimodal Signaling Affects Prompting Behaviour, Tianyi Zhang, Colin Au Yeung, Emily Aurelia, Yuki Onishi, Neil Chulpongsatorn, Jiannan Li, Anthony Tang

Research Collection School Of Computing and Information Systems

Current voice agents wait for a user to complete their verbal instruction before responding; yet, this is misaligned with how humans engage in everyday conversational interaction, where interlocutors use multimodal signaling (e.g. nodding, grunting, or looking at referred to objects) to ensure conversational grounding. We designed an embodied VR agent that exhibits multimodal signaling behaviors in response to situated prompts, by turning its head, or by visually highlighting objects being discussed or referred to. We explore how people prompt this agent to design and manipulate the objects in a VR scene. Through a Wizard of Oz study, we found that …

Quantitative Runtime Monitoring Of Ethereum Transaction Attacks, Xinyao Xu, Ziyu Mao, Jianzhong Su, Xingwei Lin, David Basin, Jun Sun, Jingyi Wang

Research Collection School Of Computing and Information Systems

The rapid growth of decentralized applications, while revolutionizing financial transactions, has created an attractive target for malicious attacks. Existing approaches to detecting attacks often rely on predefined rules or simplistic and overly-specialized models, which lack the flexibility to handle the wide spectrum of diverse and dynamically changing attack types. To address this challenge, we present a general and extensible framework, MoE (Monitoring Ethereum), that leverages runtime verification to detect a wide range of attacks on Ethereum. MoE features an expressive attack modeling language, based on Metric First-order Temporal Logic (MFOTL), that can formalize a wide range of attacks. We integrate …

Integrating Path Selection For Symbolic Execution And Variable Selection For Constraint Solving, Shunkai Zhu, Jun Sun, Jingyi Wang, Zhenbang Chen, Peng Cheng

Research Collection School Of Computing and Information Systems

Symbolic execution is a powerful technique that can accurately synthesize program inputs for program testing through constraint solving. Applying symbolic execution effectively means that we must solve two searching problems efficiently. One is to search through the many program paths and the other is, given a particular path condition, to search through the numerous variable assignments to identify one satisfying solution. With few exceptions, existing symbolic execution engines treat constraint solvers as black boxes. As a result, the two searches are completely separated, which results in much redundancy (i.e., the same variable assignments may be tried for solving many program …

Tensorjsfuzz: Effective Testing Of Web-Based Deep Learning Frameworks Via Input-Constraint Extraction, Lili Quan, Xiaofei Xie, Qianyu Guo, Lingxiao Jiang, Sen Chen, Junjie Wang, Xiaohong Li

Research Collection School Of Computing and Information Systems

The 2025 ACM Web Conference (WWW '25) took place from April 28 to May 2, 2025, in the Sydney Convention & Exhibition Centre, Australia. Its logo, featuring the Sydney Harbour Bridge, symbolizes the core "connecting" function of the Web. Formerly known as the International World Wide Web Conference (WWW), this event originated at CERN in 1994 and has long served as the premier venue for presenting and discussing research, development, standards, and applications related to the Web.The 2025 ACM Web Conference (WWW'25) took place from April 28 to May 2, 2025, in the Sydney Convention & Exhibition Centre, Australia. Its …

Decictor: Towards Evaluating The Robustness Of Decision-Making In Autonomous Driving Systems, Mingfei Cheng, Xiaofei Xie, Yuan Zhou, Junjie Wang, Guozhu Meng, Kairui Yang

Research Collection School Of Computing and Information Systems

Autonomous Driving System (ADS) testing is crucial in ADS development, with the current primary focus being on safety. However, the evaluation of non-safety-critical performance, particularly the ADS's ability to make optimal decisions and produce optimal paths for autonomous vehicles (AVs), is also vital to ensure the intelligence and reduce risks of AVs. Currently, there is little work dedicated to assessing the robustness of ADSs' path-planning decisions (PPDs), i.e., whether an ADS can maintain the optimal PPD after an insignificant change in the environment. The key challenges include the lack of clear oracles for assessing PPD optimality and the difficulty in …

Specgen: Automated Generation Of Formal Program Specifications Via Large Language Models, Lezhi Ma, Shangqing Liu, Yi Li, Xiaofei Xie, Lei Bu

Research Collection School Of Computing and Information Systems

In the software development process, formal program specifications play a crucial role in various stages, including requirement analysis, software testing, and verification. However, manually crafting formal program specifications is rather difficult, making the job time-consuming and labor-intensive. Moreover, it is even more challenging to write specifications that correctly and comprehensively describe the semantics of complex programs. To reduce the burden on software developers, automated specification generation methods have emerged. However, existing methods usually rely on predefined templates or grammar, making them struggle to accurately describe the behavior and functionality of complex real-world programs. To tackle this challenge, we introduce SpecGen, …

Oscar: Object Status And Contextual Awareness For Recipes To Support Non-Visual Cooking, Franklin Mingzhe Li, Kaitlyn Ng, Bin Zhu, Patrick Carrington

Research Collection School Of Computing and Information Systems

Following recipes while cooking is an important but difficult task for visually impaired individuals. We developed OSCAR (Object Status Context Awareness for Recipes), a novel approach that provides recipe progress tracking and context-aware feedback on the completion of cooking tasks through tracking object statuses. OSCAR leverages both Large-Language Models (LLMs) and Vision-Language Models (VLMs) to manipulate recipe steps, extract object status information, align visual frames with object status, and provide cooking progress tracking log. We evaluated OSCAR’s recipe following functionality using 173 YouTube cooking videos and 12 real-world non-visual cooking videos to demonstrate OSCAR’s capability to track cooking steps and …

Query Understanding In Llm-Based Conversational Information Seeking, Yifei Yuan, Zahra Abbasiantaeb, Yang Deng, Mohammad Aliannejadi

Research Collection School Of Computing and Information Systems

Query understanding in Conversational Information Seeking (CIS) involves accurately interpreting user intent through context-aware interactions. This includes resolving ambiguities, refining queries, and adapting to evolving information needs. Large Language Models (LLMs) enhance this process by interpreting nuanced language and adapting dynamically, improving the relevance and precision of search results in real-time. In this tutorial, we explore advanced techniques to enhance query understanding in LLM-based CIS systems. We delve into LLM-driven methods for developing robust evaluation metrics to assess query understanding quality in multiturn interactions, strategies for building more interactive systems, and applications like proactive query management and query reformulation. We …