Digital Commons @ Florida Atlantic
Not a member yet
400 research outputs found
Sort by
ROBOTIC MONITORING SYSTEM FOR COMPLEX AQUATIC ENVIRONMENTS
No full textPersistent monitoring in an aquatic environment is critical in many scientific and industrial applications. In-situ sensors that record water quality metrics, like dissolved oxygen, are essential in many aquaculture applications. Existing monitoring solutions can be cost prohibitive, difficult to maintain, and are typically deployed in fixed locations. This limits the usefulness of such systems in complex aquatic environments like coastal zones or pond aquaculture farms where bodies of water can be isolated from each other. In this dissertation research, an autonomous data collection framework is proposed to provide low-cost water quality monitoring through the combination of a variety of sensors and sensing platforms, such as waterproof UAVs and truck-based systems, that are optimized for in-situ sampling across disconnected bodies of water. Sensors and their platforms are connected to create an Internet of Things (IoT) sensor network for such environments. Sensor data collected from the system is transmitted and stored in a cloud database that provides users access to collected information via a web interface and text message alert system. To this end, this dissertation presents the methodology for designing optimal platforms, sensors, and their integration
THE CURRENT OF ANXIETIES: GHOSTS AND CAPITAL FOR THE LIMINAL BRIDE IN HIROKO OYAMADA’S THE HOLE
No full textThis thesis offers a reading of Hiroko Oyamada’s The Hole as discourse surrounding modern neoliberalism’s dissolution and transformation of boundaries between socio-political domains, such as the public/private, and its consequent effects on individual subjectivity and interpersonal relationships. Within the novella’s depictions of the mundane, I also seek to argue that Oyamada’s employment of the liminal, through distortions of space and time, is used to depict the precarity of marginalization as an always present force due to the pervasiveness of capitalism, but purposefully hidden to maintain security in illusion
REGIONAL DETERMINANTS OF BANKRUPTCY RATES IN THE U.S.
No full textThis study investigates the associations of state-level regional economic indicators about the labor market and bankruptcy rates in United States. This research aims to fill a gap in the bankruptcy literature by analyzing how these factors are related to corporate and individual bankruptcy filings. The unemployment rate is the primary labor-market predictor of financial distress; I argue that a comprehensive analysis incorporating broader labor market statistics provides a more complete picture of bankruptcy trends.
Employing state-level panel data, the study extends prior research that mainly utilized cross-sectional or time-series data. This approach provides a detailed examination of how labor market indicators and bankruptcy rates have evolved across the United States. Using state-level labor-market variables and the rate of bankruptcy filings, my findings indicate that the employment-population ratio and labor force participation rate offer valuable insights into the determinants of bankruptcy rates when combined with the unemployment rate. By systematically exploring the literature, employing robust research methodologies, and analyzing state-level panel data, this essay aims to contribute to the academic and practical understanding of the economic drivers of bankruptcy. It advocates for a nuanced view of economic indicators to better predict and manage financial instability, positioning its findings within the broader discourse on economic resilience and bankruptcy\u27s role in the United States economy
DEEP LEARNING FOR UNDERWATER MANATEE COUNTING AND TRACKING
No full textThe Florida manatee (Trichechus manatus latirostris), a threatened marine mammal often found in coastal waters, require accurate monitoring to inform conservation strategies. However, traditional counting methods are hindered by observer bias and surface distortions. Recent advances in artificial intelligence (AI) address these challenges using captured aerial footage of manatees, demonstrating the potential of artificial intelligence for automating manatee counting. Still, various challenges arise hindering accurate counts.
Building on these studies, this work advances artificial intelligence with underwater footage, instance segmentation, data augmentation, and pseudo-labeling to automate manatee counting in data-scarce environments. Three experiments were conducted using 10%, 60%, and 90% labeled training data to evaluate model performance under varying levels of data availability. Across all experiments, the proposed methods outperformed the previous state-of-the-art method, with reductions over 65% in both MAE and MASE, showing that pseudo-labeling effectively mitigates data scarcity
ESSAYS ON ENTREPRENEURIAL FUNDING IN THE ERA OF FINTECH INNOVATION
Full text linkEquity crowdfunding introduces a contemporary framework for the longstanding practice of aggregating capital from diverse groups of investors, ranging from small individual contributors to large, accredited participants. Specialized digital platforms facilitate this capital assembly process by intermediating transactions between entrepreneurs seeking financing and investors, who receive financial securities in return for their contributions. This fundraising mechanism gained prominence in the United States following the implementation of Title III of the Jumpstart Our Business Startups (JOBS) Act in 2016, which established regulatory frameworks permitting companies to publicly solicit investments online from diverse investor pools.
While equity crowdfunding has expanded capital accessibility, a fundamental challenge lies in the inherent uncertainty surrounding early-stage ventures, which hinders investors\u27 ability to assess risk and determine value. This dissertation employs a two-essay framework using comprehensive data from U.S.-regulated equity crowdfunding campaigns (May 2016-December 2023) to explore how institutional choices serve as credible signals to investors, reducing informational gaps and enhancing outcomes.
The first essay examines incorporation location as a legitimacy signal for small firms. Specifically, it investigates whether Delaware incorporation—with its well-developed legal framework and efficient dispute resolution mechanisms—enhances credibility among investors. Since crowdfunding firms typically lack operational history and have limited mechanisms to mitigate agency concerns, tying themselves to a jurisdiction with strong institutional underpinnings may signal quality to investors. The corresponding analysis reveals that Delaware-incorporated firms achieve superior performance in crowdfunding campaigns, suggesting that legal choices may play a critical role in bridging the startup-investor information gap.
The second essay examines the contrasting signaling effects of equity versus debt offerings. This analysis explores how issuing equity securities, which align founder interests with stockholders, affects campaign success and follow-on funding outcomes. The findings indicate that investors and follow-on investors favor equity securities, specifically preferred equity, more substantially than debt securities, which carry repayment obligations and constrain the operational flexibility of the crowdfunding firm.
Together, these essays contribute to understanding how institutional signals mitigate information asymmetry in entrepreneurial finance, offering implications for entrepreneurs seeking capital, platforms facilitating transactions, and policymakers designing regulatory frameworks in the evolving fintech landscape
EXPLORING A MODERN DEEP LEARNING TECHNIQUE FOR WETLAND MAPPING AND MONITORING USING WORLDVIEW-2 SATELLITE PRODUCTS
Full text linkWetlands play a significant role in the world’s hydrology, climate, and biodiversity. Even with the benefits and values wetlands provide to the environment, they have been undergoing loss and degradation due to natural and anthropogenic processes. To protect wetlands from loss and degradation and to restore their function, it is essential to develop a sustainable wetland monitoring system. One of the key elements of a wetland monitoring system is wetland mapping. This dissertation research developed an object-based deep learning protocol for mapping heterogeneous wetlands with many communities from a high-resolution WV-2 satellite image. To test this developed protocol, an object-based machine learning ensemble approach was selected as a benchmark for comparison. To effectively apply the developed protocol, feature selection techniques were applied, optimal spectral and spatial features were identified, and the benefit of four additional bands of WV-2 products were evaluated. The study also applied a post classification change detection technique to delineate the change between 2017 and 2021. The developed object-based deep learning protocol has been proven superior to the object-based machine learning ensemble approach. The feed-forward neural network (FNN) deep learning classifier achieved an overall accuracy of 91.2% and 88.6% for 2017 and 2021 imagery, respectively. On the other hand, the ensemble analysis approach achieved an overall accuracy of 87.8% and 85.5% for 2017 and 2021 imagery, respectively. The FNN improved (\u3e3%) the classification accuracy compared to the ensemble analysis, and the difference between classification results was statistically significant. The deep learning classifier not only increased the overall accuracy, but it also helped identify minor communities more accurately than the ensemble analysis technique. The additional four bands, object-based texture measures, and NDVI values of WV-2 satellite imagery showed the potential to map heterogeneous wetlands with many communities. The change map provided valuable insights into temporal changes in wetlands, which can aid in the formulation of adaptive management strategies. Exploration of automated/semi-automated deep learning methods contributed by this dissertation research will not only advance modern deep learning in wetland applications, but also assist with regional land managers to make efficient decisions by generating timely map products
PRESS PASS: A MIDDLE EAST MEMOIR OF IDENTITY AND EMPATHY
No full textThis is a work of memoir and social commentary that touches on issues of identity, race, religion, gender, and indigeneity. The author, an American Jewish journalist who spent 20 years covering the Middle East and other parts of the Islamic world, examines her experiences of passing (i.e. as a local, native, or indigenous person) in the countries she covered, and questions how this impacted her understanding of the region and her empathy for the people she wrote about as a foreign correspondent. The memoir invites readers on a journey that extends through multiple countries in a time of war, most notably Iraq, Afghanistan and Israel-Palestine. The style of writing draws on nonlinear storytelling techniques, lyric essay, and a weaving of past and present-tense narratives. The work was influenced by contemporary literature on passing, liminality, racism, colorism, feminism, Islamophobia, and antisemitism, as well as ancient Jewish texts in which characters pass as others as a means of survival
CLINICAL VALIDATION OF DELIVERY ANALYSIS SOFTWARE FOR PATIENT-SPECIFIC QUALITY ASSURANCE OF RADIXACT (TOMOTHERAPY) TREATMENT
Full text linkThe primary objective of this study was to evaluate the patient-specific quality assurance (PSQA) performance for various disease sites treated with the Radixact machine using Gamma (γ) passing rate analysis (%GP) as per AAPM TG-218.
This study analyzed 22 PSQA plans executed on a Radixact machine using ArcCheck phantom and delivery analysis (DA) software. The %GP was calculated using a 3% dose difference (DD), a 2 mm distance-to-agreement (DTA), and a 10% threshold. A passing criterion of ≥95% was applied for PSQA evaluation.
The %GP was analyzed for all sites. The %GP (2D) on the ArcCheck phantom mean γ passing rate with variability was (97.12±2.63%), while the %GP (3D) rate for DA was 100%.
For DA, the %GP was 100% because it only considers the MLC-LOTs patterns, not plan parameters, to compute the delivered versus planned dose. Investigations should be required to recalibrate the output of the MVCT diode array
AN ANALYSIS OF DOCTOR OF PHILOSOPHY TIME-TO-DEGREE PREDICTORS
Full text linkDoctoral-granting institutions invest time and resources to support the success of their Doctor of Philosophy (PhD) students. Despite these efforts, doctoral attrition rates remain high. One measure used to understand doctoral attrition and completion rates is time-to-degree (TTD). In 2020, the median TTD from the start of graduate school increased to 7.5 years (National Center for Science and Engineering Statistics [NCSES], 2020b), highlighting the need for continued analysis of factors affecting TTD. This study used data from the 2021 Survey of Earned Doctorates (SED; NCSES, 2021a) to provide a more recent analysis of the relationship between TTD and both student and institutional characteristics.
Astin’s Input-Environment-Output (I-E-O) model served as the conceptual framework and helped guide the variables included in this study (Astin, 1970a; Astin & Antonio, 2012). The input variables included student characteristics (i.e., gender, race/ethnicity, age, marital status, dependents, citizenship status), while the environment variables consisted of institutional characteristics (i.e., Carnegie Classification, institutional control, field of study, financial support). TTD was the output or dependent variable. Both univariate and multivariate analyses were conducted to examine the relationships between the variables and TTD. The sample for this study included 49,358 PhD recipients from 448 public and private institutions across the United States.
The univariate analyses revealed statistically significant differences; however, the practical significance of all student and institutional characteristics was relatively small. When student and institutional characteristics were combined, the multiple linear regression showed that the model was significant, accounting for approximately 9.9% of the variance in TTD. When the subset of student characteristics was added to the model that included institutional characteristics, the model was significant, with a change in R² of 1.4%. Similarly, when the subset of institutional characteristics was added to the model that included student characteristics, the model was significant, with a change in R² of 5.3%. This indicates that institutional factors contributed more to the model than student factors.
Although this study confirmed the relationship between TTD and both student and institutional characteristics, these factors had a relatively small influence, whether individually or combined. Due to the low predictive power, other, more nuanced factors may provide better insights into what drives PhD degree completion times. These findings should encourage institutions and policymakers to adopt more holistic strategies to foster doctoral student success
ENTREPRENEUR\u27S FAMILY STATUS, STRATEGY, PERFORMANCE
Full text linkThis research aims to illuminate the relationship between marital status and entrepreneurial performance among entrepreneurs, focusing on the implications for female entrepreneurs. While existing studies have predominantly examined gender-related challenges, such as disparities in access to financial and social capital, the role of marital status has received limited attention. Drawing on Becker’s seminal work (1965, 1974) on time allocation and human capital investment, this study argues that entrepreneurs face distinct risks and resource constraints compared to employees, who benefit from organizational risk-sharing and stable income. In contrast, entrepreneurs bear the full burden of market uncertainty and opportunity costs, making personal life factors, especially marital support, critical determinants of success.
Leveraging comprehensive data from the Entrepreneurship in the Population (EPOP) surveys conducted in 2022, 2023, and 2024, the study employs advanced econometric techniques, including ordered logistic regression analyses, to test two primary hypotheses. The first hypothesis posits that marital status significantly influences entrepreneurial performance, with married or living with a partner entrepreneur achieving higher profitability and revenue due to shared household resources, risk-sharing, and reduced opportunity costs. The second hypothesis suggests that gender moderates this relationship, as female entrepreneurs encounter compounded challenges, such as intensified work-family conflict and restricted access to capital, which may diminish the benefits of marital support.
The empirical results robustly support the first hypothesis. Across multiple model specifications ranging from baseline models to fully augmented specifications with entrepreneurial and firm-level controls, the findings consistently reveal that entrepreneurs who are single or widowed/divorced/separated report significantly lower profitability and occupy lower revenue tiers compared to their married or living with a partner counterpart. These patterns persist in robustness tests across subsamples defined by legal ownership structures, further underscoring the critical role of spousal support in fostering business success. Although the evidence for the second hypothesis is more preliminary, the consistently negative coefficients associated with the Female indicator suggest that gender-based constraints exacerbate performance disadvantages.
By integrating Becker’s time allocation theory with Social Role Theory, this research bridges a critical gap in the literature and offers actionable policy recommendations. The findings highlight the need for targeted interventions such as subsidized childcare, flexible work arrangements, and enhanced access to credit to mitigate work-family conflict and promote gender equity in entrepreneurship. Ultimately, these insights advance both theoretical debates and practical strategies for fostering a more inclusive entrepreneurial ecosystem