University of Illinois Urbana-Champaign
Illinois Digital Environment for Access to Learning and Scholarship RepositoryNot a member yet
123813 research outputs found
Sort by
Quantifying soil compaction effects on soil macropores and corn root morphology through image analysis
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Beau Barber, accepted the attached license on 2025-04-24 at 19:57.The student, Beau Barber, submitted this Dissertation for approval on 2025-04-24 at 20:04.This Dissertation was approved for publication on 2025-05-02 at 07:03.DSpace SAF Submission Ingestion Package generated from Vireo submission #21969 on 2025-10-19 at 18:19:09Excessive soil compaction is among the most significant influences on soil quality and crop production worldwide. The source of excessive soil compaction is primarily due to the increase in agricultural machinery mass and changes in field management strategies, which have substantially increased agricultural productivity over the last century. The prevalence of excessive agricultural traffic-induced compaction has led to the development of a broad field of research attempting to understand the mechanisms of soil compaction in agriculture and how it could be managed, without making significant sacrifices to modern farming equipment and practices that have led to the high yielding production that civilization enjoys. Many methodologies exist to both analyze and manage the impacts of soil compaction, each with specific implications and conclusions that better inform the scientific understanding of soil behavior and its impact on agriculture. The goal of the research within this manuscript was to evaluate the soil compaction induced by different agricultural machine systems and field management decisions and the resulting impacts on soil quality and root morphology as they pertain to the soil compaction induced by these practices. This research involved three case studies that evaluate the impact of soil compaction induced by different agricultural practices on soil pore and corn root morphology via image analysis
An ontology-guided, language model-assisted approach for theme-specific information extraction
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Jinfeng Xiao, accepted the attached license on 2025-04-25 at 13:12.The student, Jinfeng Xiao, submitted this Dissertation for approval on 2025-04-25 at 13:37.This Dissertation was approved for publication on 2025-04-25 at 16:40.DSpace SAF Submission Ingestion Package generated from Vireo submission #21985 on 2025-10-19 at 18:19:17In an era of information explosion, text data are generated every second so fast that digesting raw texts becomes increasingly time-consuming. The research field of text mining aims at automatically extracting structured knowledge from massive unstructured text data. An essential step in the process of converting text to knowledge is information extraction (IE), which aims at extracting information about entities and relations from text. In recent years, the development of large language models (LLMs) has shed light on new possibilities of performing better IE under label-sparse settings that are hard, if not impossible, for traditional methods due to the lack of human supervision that is necessary for model training. On the other hand, LLM-assisted domain-specific IE applications often face long-tail challenges, which refer to suboptimal performance due to the fact that many important signals (e.g., entities and relations) in the target domain are sparse in the pre-training corpora and are thus not handled well by pre-trained LLMs. Such long-tail problems become more severe when we move one step further from domains into highly specialized themes. My dissertation research explores the utilization of theme-specific entity type ontologies to improve the long-tail robustness of LLMs for IE tasks. I utilize such ontologies to 1) retrieve theme-relevant documents from large corpora based on the distribution of query entities and relations on the ontology, 2) find entities from large corpora that have fine-grained entity types relevant to the query entities, 3) retrieve relevant external, non-parametric knowledge in a structured manner to augment LLM queries, and 4) organize the internal, parametric knowledge of LLMs into relation structures for robust open relation extraction. While the 1st and 3rd components focus more on information retrieval (IR), all the components together form an end-to-end framework for corpus-level theme-specific IE. Extensive experiments demonstrate that my solutions achieve state-of-the-art performance with remarkable long-tail robustness. One more notable advantage of my approach is that it works purely at the LLM inference stage. No extra pre-training or post-training is required. This means my approach works without additional annotation efforts or heavy resource consumption. All my methods are easy to deploy on a single graphics processing unit (GPU) with any recent LLMs, and thus, my solutions are highly accessible to ordinary people. In addition, my design essentially separates data from models, which ensures there is technically zero risk of leaking private user data. Thus, my approach is highly suitable for sensitive domain applications such as medical care and legal advising
Essays on competition, innovation, and the interplay of corporate and business strategies
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01The student, Valerie Isakova, accepted the attached license on 2025-03-04 at 10:53.The student, Valerie Isakova, submitted this Dissertation for approval on 2025-03-04 at 11:04.This Dissertation was approved for publication on 2025-03-06 at 12:50.DSpace SAF Submission Ingestion Package generated from Vireo submission #21659 on 2025-10-19 at 19:14:21This dissertation unpacks the complex relationship between the multidimensional concept of competition and a firm’s innovation, such as new product introductions. New product introductions are not only a key driver of a firm’s growth and profitability but also a firm’s strategic response to competitive pressures. Existing literature on the relationship between competition and a firm’s new product introductions primarily considered competition with rivals in product markets. However, whether a firm increases or reduces its new product introductions in response to other distinct dimensions of competition—such as competition with rivals for critical resources in strategic factor markets, competition within a firm’s product portfolio in a given business (product category), and competition across distinct businesses (product categories) within a firm—is not clear and remains the topic of ongoing inquiry in strategy research. The dissertation essays seek to contribute to strategy research by examining the above-mentioned under-researched dimensions of competition. The first essay, “Toward a synthesis of corporate strategy and business strategy: New product introductions and product line extensions by diversified firms” (Chapter 2), focuses on cross-business strategic interdependencies generated by the interplay of specific corporate and business strategies, such as corporate diversification and vertical product line extension. Unlike prior literature that considered economies of scope arising from resource relatedness a primary strategic interdependence that increases the economic profitability of diversified firms, this essay puts forward the business strategies that a diversified firm implements in its businesses as another crucial factor that affects its economic performance. Specifically, the study discusses how diversified firms use the business strategy of vertical product line extension to increase their revenues. The essay reveals that negative strategic interdependencies across a diversified firm’s businesses, such as competition across distinct businesses (product categories) within a diversified firm, affect its choice of the business strategy and limit its ability to use the business strategy of vertical product line extension in its businesses (product categories). The second essay, “How product similarity and rivals’ product launches and withdrawals affect a firm’s new product introductions” (Chapter 3), examines how competition among a firm’s products within a given business (product category), such as similarity among a firm’s products, and competition with rivals in product markets, such as rivals’ new product launches and withdrawals, affect a firm’s new product introductions. The study reveals that competition among a firm’s products and competition with rivals in product markets suppress a firm’s new product introductions. The third essay, “Competition for critical resources and a firm’s product innovation and product deployment strategy” (Chapter 4), focuses on competition with rivals for critical resources, such as rivals’ relative strength in expanding their geographic coverage. The study highlights that a firm responds to competition with rivals for critical resources by renewing its product portfolio. Specifically, the firm discontinues its maturing products and introduces new ones
Neurocognitive function and influence of diet among persons with multiple sclerosis
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01The student, Shelby Martell, accepted the attached license on 2025-03-14 at 10:01.The student, Shelby Martell, submitted this Dissertation for approval on 2025-03-14 at 10:20.This Dissertation was approved for publication on 2025-04-01 at 12:00.DSpace SAF Submission Ingestion Package generated from Vireo submission #21676 on 2025-10-19 at 19:14:25Objectives Multiple sclerosis (MS) is an inflammatory and demyelinating disease that disrupts the central nervous system causing a range of physical and cognitive impairments. Current treatments for MS have limited success addressing cognitive, motor, and visual impairments. Thus, there is a growing interest and evidence in lifestyle factors as complementary approaches to support MS management. Nutrition is of key interest for people with MS however, dietary recommendations for improving MS outcomes remain unknown. This dissertation aimed to address these gaps by exploring the interrelationships between nutrition, cognitive function, and MS disease markers in persons with MS. Aim 1 investigated the effects of lutein supplementation on carotenoid status and cognitive function in persons with MS. Aim 2 investigated how the lateralized readiness potential (LRP), an index of motor response and activation in the brain, differs in MS and HC as well as the relationships. Aim 3 assessed relationships between retinal morphologic markers of MS, cognitive performance, event-related potentials (ERP), and optic neuritis (ON) in persons with MS. Lastly, aim 4 investigated the feasibility of a Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) diet intervention in MS and the potential benefits for cognitive function. Methods We utilized both clinical pilot intervention trials and cross-sectional studies to explore cognitive function and nutrition in MS. Aim 1 was a single-blind placebo controlled RCT of 20 mg lutein supplementation in persons with MS (n = 21; active n = 12, placebo n = 9). Carotenoid status was evaluated utilizing skin carotenoids via reflection spectroscopy, macular pigment optical density (MPOD) via heterochromatic flicker photometry, and serum lutein. Cognitive function was assessed via a modified Eriksen flanker task with simultaneous EEG and the symbol digits modalities test (SDMT). Aim 2 utilized secondary data analysis in persons with MS (n = 53) and age-sex matched HC (n = 53) with flanker EEG data to explore the lateralized readiness potential (LRP) both in relation to the stimulus (LRP-S) and response (LRP-R) as well as flanker accuracy and reaction time. Aim 3 involved secondary data in persons with MS (n = 53) with optical coherence tomography (OCT) scans to assess three markers of retinal morphology: retinal nerve fiber layer (RNFL), total macular volume (TMV), and granule cell inter plexiform layer (GCIPL). Additionally, flanker EEG data was utilized with cognitive (P3, N2, and ERN) and motor (LRP-S and LRP-R) ERP components. Finally Aim 4, was a telehealth 12-week pilot RCT in persons with MS (n = 31) exploring feasibility and effects of MIND diet (n = 16) compared to control diet (n = 15) on cognitive function. Participants consumed a ready-to-eat meal per day and followed diet guidelines. Feasibility was assessed through descriptive analysis of retention, attrition, and compliance rates. Additionally, the effects of the intervention on diet indices were calculated from the dietary history questionnaire and adherence surveys. Cognitive function was assessed via the remote testing CANTAB cognitive battery. Results Analyses conducted for aim 1, after adjusting for baseline serum lutein and age, revealed significant group by time interactions for MPOD, skin carotenoids, and serum whereby the treatment group improved in all carotenoid outcomes. In the treatment group increase in MPOD was positively associated with incongruent accuracy and spatial memory. Aim 2 revealed that participants with MS had delayed LRP-S latency and reduced LRP-R amplitudes compared to HC. In Aim 3, significant relationships were observed with RNFL and accuracy, reaction time, P3 congruent latency, LRP-S amplitude and ERN peak latency. Additionally, GCIPL was significantly related to accuracy, SDMT score, LRP-R amplitude, LRP-S amplitude and LRP-S latency. Lastly TMV was significantly predictive of N2 amplitude. Finally, for the 12-week intervention conducted in Aim 4, the retention rate of 87.5% with an attrition rate of 12.5%. Compliance rates for consumption of the daily meal were high with overall compliance of 91.1% with the MIND group having 89.7% compliance and 92.6% compliance in the control diet. There was a significant time effects whereby HEI improved over time and only the MIND group had significant increases in MIND composite. Significant time effects were observed whereby post-test memory scores and attention reaction time improved. No other significant effects were observed for the cognitive outcomes. Conclusions These works provide evidence supporting the potential neuroprotective benefits of nutrition in MS and highlight neuroelectric indices of cognition and motor processes as potential biomarkers for MS. Specifically, Aim 1 demonstrated that four months of lutein supplementation significantly increased carotenoid status, with increases in macular pigment linked to improvements in attentional inhibition and spatial memory. Aim 2 revealed impairments in delayed pre-motor response selection and reduced neural activation for motor response initiation in MS. Aim 3 revealed novel relationships between proxy markers of white and gray brain matter and cognitive, motor, and error processes in MS. Finally, the remote pilot trial in Aim 4 demonstrated high feasibility and acceptability. While the MIND diet did not show greater cognitive benefits compared to the control diet, the findings support the need for larger fully powered RCTs to explore dietary patterns and cognition in MS. Collectively, this dissertation work informs future interventions and larger efficacy trials, offering valuable insights for researchers and clinicians to design interventions to provide neurocognitive benefits in persons with MS
Integrating cyberGIS and big data for scalable spatial accessibility analysis
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01The student, Alexander Michels, accepted the attached license on 2025-04-07 at 10:10.The student, Alexander Michels, submitted this Dissertation for approval on 2025-04-07 at 10:15.This Dissertation was approved for publication on 2025-04-07 at 12:02.DSpace SAF Submission Ingestion Package generated from Vireo submission #21717 on 2025-10-19 at 19:14:30Spatial accessibility describes the uneven spatial distribution of access to vital resources and services like healthcare and critical infrastructure. Access is typically measured using travel time and ratios of supply-to-demand at the population-level to determine a level of access for each spatial unit such as a census tract. While the growth in geospatial big data and advances in cyberinfrastructure-based geographic information science and systems (cyberGIS) have widened the possibilities for spatial accessibility analysis, providing new data sources and research capabilities, the spatial accessibility literature has lagged behind this trend. This dissertation details methodological advancements in the field of cyberGIS-enabled spatial accessibility with three main research objectives: (1) the development of algorithms and methods for scalable travel time and spatial accessibility analyses, (2) the design of innovative approaches to harnessing emerging mobility data sources, and (3) the application of machine learning techniques to emerging big data sources to enhance the comprehension and approximation of travel time within spatial accessibility analyses. Our results underscore the efficacy of cyberGIS-enabled spatial accessibility through novel applications to real-world datasets and case studies. This approach enables researchers and decision-makers to analyze accessibility at larger scales, finer granularities, and with higher accuracy than what is currently achievable with existing methods
Resilient Yet Vulnerable: Understanding the Psychosocial Experiences of Black Women First-Generation College Students in STEM
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01The student, Destiny Williams-Dobosz, accepted the attached license on 2025-04-18 at 14:54.The student, Destiny Williams-Dobosz, submitted this Dissertation for approval on 2025-04-18 at 15:16.This Dissertation was approved for publication on 2025-04-21 at 16:23.DSpace SAF Submission Ingestion Package generated from Vireo submission #21837 on 2025-10-19 at 19:15:10The ontological experiences (realities and ways of being) and epistemologies (ways of knowing) of underserved and marginalized populations are not one-dimensional. This research aimed to explore how Black women who are first-generation college students (FGCSs) navigate Science, Technology, Engineering, and Mathematics (STEM). As an educational psychologist who researches help-seeking and learning in STEM for numerically underrepresented and minoritized students, I became frustrated reading deficit-based approaches to research. This frustration brought me to the quote, “Black women are simultaneously vulnerable and resilient in the face of systematic inequality” (Evans-Winters, 2019). From this liminal space, I began a research journey to understand the both-and tension of resilience and vulnerability for Black women FGCSs in STEM. First, I employed autoethnographic storytelling to uncover my developmental trajectory toward graduate studies. As a Black woman, FGCS, and as someone who studied chemistry as an undergraduate, I needed to grapple with my own stories to analyze other stories responsibly. Next, I conducted a single-case interpretative phenomenological analysis of Jane (pseudonym), a Black woman FGCS in computer science, to explore the risk contributors and protective factors to her STEM pursuit. Last, I conducted a cross-case interpretative phenomenological analysis of four Black women FGCSs in STEM to investigate their net vulnerability, stress engagements, and coping mechanisms when seeking academic help. Ultimately, each study reported in this dissertation presents layered and practical insights for instructors, administrators, and students. To address any educational problem effectively, we must first know what is happening, what is working, what is not working, and for whom it is not working
Advanced assisted reproductive technology programs in developing countries in the tropics: Factors affecting establishment and success
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01The student, Elizabeth Bangert, accepted the attached license on 2025-04-24 at 10:06.The student, Elizabeth Bangert, submitted this Thesis for approval on 2025-04-24 at 10:07.This Thesis was approved for publication on 2025-05-06 at 11:41.DSpace SAF Submission Ingestion Package generated from Vireo submission #21926 on 2025-10-19 at 19:16:02Despite global increases in food production, chronic food insecurity persists in tropical developing countries, with projections indicating nearly 600 million people will remain undernourished by 2030. This research addresses this critical challenge by investigating Assisted Reproductive Technologies (ARTs) as an innovative strategy to enhance livestock productivity in Tanzania, where 31.8% of children under 5 are malnourished and agricultural productivity is severely constrained by climate variability and insufficient infrastructure. A case study in the Mpwapwa Region of Tanzania demonstrated the potential and challenges of implementing advanced ARTs in a developing country context. Of 110 potential bovine embryo recipients, only one pregnancy was confirmed, highlighting the complex barriers to genetic improvement. Key findings revealed that successful ART programs require a holistic approach addressing technological, educational, and systemic challenges. Overall, this thesis evaluates an effort to implement an ART program in a developing country and suggests ways to make future efforts go more smoothly. Future directions should include developing a comprehensive roadmap for developing countries seeking to leverage advanced reproductive technologies. This research offers critical insights into efforts to enhance livestock productivity and, consequently, food security in challenging tropical environments
Technoeconomic analysis and life cycle assessment of electrochemical processes for value-added chemicals
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01The student, Adam Sibal, accepted the attached license on 2025-04-25 at 09:39.The student, Adam Sibal, submitted this Dissertation for approval on 2025-04-25 at 10:12.This Dissertation was approved for publication on 2025-04-29 at 17:11.DSpace SAF Submission Ingestion Package generated from Vireo submission #21975 on 2025-10-19 at 19:16:18The chemical industry is a major contributor to global greenhouse gas emissions, with ethylene oxide and propylene oxide production ranking among the most energy- and emissions-intensive processes. Additionally, the transportation sector, particularly diesel-powered heavy-duty vehicles, remains a significant source of CO₂ emissions that could be alleviated with increased biodiesel adoption. This dissertation explores electrochemical oxidation as a potential decarbonization pathway for both industrial chemical production and biodiesel co-product utilization, providing a comprehensive assessment of its technoeconomic feasibility, environmental trade-offs, and policy implications. A technoeconomic analysis was conducted on the direct electrochemical epoxidation of ethylene and propylene, establishing experimental benchmarks for competitiveness with current thermochemical processes. The findings indicate that while electrochemical oxidation is technically feasible, achieving cost competitiveness requires advancements in gas diffusion electrode reactor design, reaction selectivity, and catalyst stability. A life-cycle assessment of electrochemical epoxidation was performed to compare its environmental impact to traditional production methods. The results show that electrification of industrial heating alone does not necessarily reduce emissions, but direct electrochemical oxidation offers a lower-carbon alternative, even under current grid conditions, with significant advantages in global warming potential, smog formation, and human health toxicity. This dissertation also examines the electrochemical oxidation of biodiesel co-produced glycerol, addressing a critical gap in existing literature, which predominantly considers 99% purity glycerol, a purity level rarely achieved in real-world biodiesel production. Through primary data collection from U.S. biodiesel producers, this research demonstrates that most facilities operate at 50–80% purity levels and confirms that electrochemical oxidation remains viable at these lower purities. This finding supports the economic and environmental potential of utilizing lower-purity glycerol streams, reducing waste and emissions while improving process efficiency. Lastly, a detailed statistical analysis was conducted to evaluate the effectiveness of state-level biodiesel policies, identifying the most impactful policy mechanisms for increasing biodiesel production and utilization. The analysis reveals that targeted incentives, infrastructure investments, and regulatory stability are key drivers of biodiesel market growth and emissions reductions, providing policymakers with data-driven insights to inform future policy design. Collectively, this research underscores that simply electrifying existing chemical production is insufficient for deep decarbonization. Instead, electrochemical oxidation presents a scalable and sustainable alternative to conventional epoxidation and glycerol valorization, offering opportunities to reduce emissions, improve resource efficiency, and eliminate hazardous byproducts
The effect of cadence-based training on time trial performance in elite wheelchair-racing athletes: A pilot study
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01The student, Annika Koester, accepted the attached license on 2025-04-29 at 12:30.The student, Annika Koester, submitted this Thesis for approval on 2025-04-29 at 12:34.This Thesis was approved for publication on 2025-05-05 at 10:14.DSpace SAF Submission Ingestion Package generated from Vireo submission #22016 on 2025-10-19 at 19:16:22This pilot study examined the impact of a cadence-based training intervention on 3-kilometer time trial performance among elite wheelchair-racing athletes. Based on insights from cycling science, we hypothesized that training at imposed cadences, ranging from very low to very high, would enhance performance outcomes in this population. This pilot study examined the impact of a cadence-based training intervention on 3-kilometer time trial performance among elite wheelchair-racing athletes. Based on insights from cycling science, we hypothesized that training at imposed cadences—ranging from very low to very high—would enhance 3-kilometer time trial performance among competitive wheelchair-racing athletes. Twelve elite wheelchair racers (6 men, 6 women; ages 18–35) completed a 21-day cadence-focused training program using wearable sensor sleeves and metronome cues to maintain individualized cadence zones during indoor roller workouts. Pre- and post-intervention 3-km time trials on a stationary roller dynamometer were conducted to measure performance changes. After the intervention, athletes' 3-km completion times significantly improved (mean improvement ≈16 s, p < .05), while their freely chosen cadence showed no significant change. Based on these results, we hypothesize that performance gains were achieved through improved propulsion efficiency and force application rather than participants altering their freely chosen cadence. These preliminary findings highlight the value of evidence-based cadence training and innovative wearable feedback tools as a potential method to enhance competitive wheelchair racing performance
Essays in applied microeconomics
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-05-01The student, Wilfried Adohinzin, accepted the attached license on 2025-04-28 at 12:45.The student, Wilfried Adohinzin, submitted this Dissertation for approval on 2025-04-28 at 12:52.This Dissertation was approved for publication on 2025-04-30 at 09:14.DSpace SAF Submission Ingestion Package generated from Vireo submission #22035 on 2025-10-19 at 19:16:24This dissertation investigates how public policies and institutional features influence labor market behavior, worker productivity, and economic outcomes across different settings. Chapter 1 investigates how job security affects teacher retention and student learning in Peru, using a nationwide education reform and administrative data. Using a regression discontinuity design, I show that exit rates from the profession decrease by 25% for teachers marginally passing the exam compared to failing, with stronger effects for higher-quality teachers, based on ex-ante value-added measures. Exploiting variation within-students across subjects, I find that students taught by permanent contract teachers experience learning gains of 0.14 standard deviations, with effects concentrated in urban areas. A difference-in-differences analysis finds that teacher value-added increases by 7.5% in schools exposed to the reform. Results highlight that test-based job security can be a useful policy tool to improve public sector productivity. Chapter 2 examines the impact of the Secure Communities (SC) program, an immigration enforcement policy, on rental markets in the United States. Using a difference-in-differences framework, I exploit the staggered rollout of SC across counties between 2008 and 2014 to estimate its effect on rental prices at the local level. Leveraging microdata from the American Community Survey (ACS), I find that SC led to a 1.3 percent statistically significant decline in average rents, with stronger effects in areas with larger Hispanic and undocumented immigrant populations. The decline in rental prices appears to be driven by reductions in immigrant housing demand, consistent with SC’s deterrent effects on undocumented populations. In contrast, I find no evidence of significant rent declines in areas with higher shares of non-Hispanic whites or Black residents. The findings highlight how policy interventions targeting specific populations can reshape urban economic conditions. Chapter 3 examines the impact of pregnancy and childbirth on labor market outcomes across 16 European countries using a pseudo-event study utilizing pre-conception data and information on month of childbirth. We document a sharp and persistent decline in mothers’ employment, with over one-third of the drop occurring during pregnancy, especially the third trimester. We also find reductions in income between the third trimester and post-childbirth, with small compensatory increases in maternity benefits. For fathers, hours worked and earnings rise briefly between conception and delivery, driven mainly by unmarried men. Monthly data and labor market outcomes during pregnancy allow us to uncover a previously overlooked pregnancy penalty for women, but a pregnancy premium for men. Together, these chapters offer new evidence on how public policy shapes labor market outcomes and contributes to broader patterns of inequality and economic opportunity