Digital Commons @ Texas A&M University-San Antonio
Not a member yet
3365 research outputs found
Sort by
Chemistry of the Groundwater in the Edwards Aquifer and Trinity Aquifer
No full textThe San Antonio area relies on groundwater from both the Edwards Aquifer and the Trinity Aquifer. We collected groundwater from three wells in the two different aquifers over various seasons to understand their chemistry. The wells, SARAMUD (Edwards Aquifer); FRP House and FRP Western (Trinity) showed changing physical and chemical parameters over the different seasons. The samples were examined for changes in pH, conductivity, and alkalinity and it showed that the alkalinity levels and conductivity are the lowest in fall and the highest in the winter. The pH levels have a different pattern, lowest in spring and highest in fall. The groundwater is more neutral in the spring than in the other seasons. These results showed that rock-water interactions influence groundwater chemistry over the various seasons
Benchmarking Robustness of Contrastive Learning Models for Medical Image-Report Retrieval
No full textMedical images and reports offer invaluable insights into patient health. The heterogeneity and complexity of these data hinder effective analysis. To bridge this gap, we investigate contrastive learning models for cross-domain retrieval, which associates medical images with their corresponding clinical reports. This study benchmarks the robustness of four state-of-the-art contrastive learning models: CLIP, CXR-RePaiR, MedCLIP, and CXR-CLIP. We introduce an occlusion retrieval task to evaluate model performance under varying levels of image corruption. Our findings reveal that all evaluated models are highly sensitive to out-of-distribution data, as evidenced by the proportional decrease in performance with increasing occlusion levels. While MedCLIP exhibits slightly more robustness, its overall performance remains significantly behind CXR-CLIP and CXR-RePaiR. CLIP, trained on a general-purpose dataset, struggles with medical image-report retrieval, highlighting the importance of domain-specific training data. The evaluation of this work suggests that more effort needs to be spent on improving the robustness of these models. By addressing these limitations, we can develop more reliable cross-domain retrieval models for medical applications
Measuring and Improving the Efficiency of Python Code Generated by LLMs Using CoT Prompting and Fine-Tuning
No full textWith the advanced AI technologies, Large Language Models (LLMs) have improved programming automation. However, LLMs often produce code with unnecessary logic, hallucinated content, and errors due to ambiguous prompts. This research measures the efficiency of Python code generated by GPT-4o-Mini, GPT-3.5-Turbo, and GPT-4-Turbo models using metrics like execution time, memory usage, and maximum memory usage, while maintaining problem-solving correctness. Using the EffiBench dataset on Google’s Vertex AI Workbench with different machine configurations, the study uses the seed parameter for consistency and optimization techniques like Chain-of-Thought (CoT) prompting and fine-tuning GPT-4o-Mini. Except for GPT-4-Turbo, the results show that CoT prompting improves efficiency metrics for GPT-4o-Mini and GPT-3.5-Turbo. GPT-4o-Mini was selected for fine-tuning due to its better results with CoT prompt and its cost-effectiveness, but fine-tuning compromises accuracy and efficiency. Overall, high-CPU machine configurations, along with GPT-4o-Mini and CoT prompting, improves the efficiency and correctness of LLM-generated code in resource-intensive scenarios
Politics in the Bedroom? Evaluation of partners’ political attitudes in romantic relationships
No full textMarital quality of life has been found to be negatively correlated with psychological distress (Gulzar et al., 2024). The purpose of this study is to determine if political affiliation based on the Triplex Scale of Political Orientation affects romantic relationship satisfaction and tolerance levels for differing political views in a monogamous couple. We hypothesize that those scoring higher in social control will score lower on relationship satisfaction and partner tolerance for differing political views than those who score higher on economic conservatism and general authoritarianism, and that scores in relationship satisfaction and partner tolerance for differing political views will be lower in couples with differing political views if the participant\u27s desired candidate loses the election, regardless of political orientation. According to the preliminary analysis of the first 97 participants, General Authoritarianism (GA) predicted lower tolerance t(52) = 2.923, p = 0.005 (p \u3c 0.05) for different political ideas, which supports the hypothesis
Characterization of Contaminants Transport in Limestone Samples from the Recharge Zone of the Edward\u27s Aquifer
No full textThe Edwards aquifer, which is a karst aquifer, is highly vulnerable to a lot of contaminants and this is because of the complex structure which makes it easy for contaminants to travel rapidly. To imitate the process by which contaminants travel into the Edwards aquifer and contaminants it, a column experiment was used to study the movement of atrazine into the karst aquifer. To quantify the concentration of atrazine before and after the experiment, a High-performance Liquid chromatography (HPLC) was used.
Also, the impact on atrazine mobility, retention, and degradation, and other parameters were determined by studying the conductivity, pH, alkalinity, and Ca2+ ions on retention and degradation were measured to illustrate the need for effective pollution control and groundwater monitoring by demonstrating how hydrogeochemical conditions impact the transport of atrazine. This experiment is paramount because it provides necessary information on how to improve the quality of the karst aquifer
Benchmarking Graph-Based Models for Real-Time IoT Anomaly Detection
No full textThe exponential growth of IoT networks introduces new security vulnerabilities that are often exploited in real time. This research poster hypothesizes that while current state-of-the-art (SOTA) real-time detection algorithms offer promising performance, they are still hindered by issues related to detection reliability, resource constraints, and adaptability in dynamic network environments. This study focuses on benchmarking graph-based models— for real-time anomaly detection using a standardized IoT attack dataset. These models leverage the structural and temporal properties of network traffic graphs to detect anomalies. We aim to evaluate their accuracy, detection latency, and computational efficiency under IoT-specific constraints. By identifying performance gaps and contextual challenges, this research will inform the development of more adaptive and lightweight graph-based detection mechanisms. The findings will contribute to advancing security in IoT networks and guiding future innovation in real-time threat detection
ASSOCIATIONS BETWEEN SOCIAL COMPARISON, STRESS AND BELONGING FOR STUDENTS AT A HISPANIC MAJORITY INSTITUTION
Get PDFHispanic students at Hispanic Majority Institutions (HMIs) experience unique social comparison processes that shape their academic self-concept, belonging, and stress levels. While research has examined Hispanic students as minorities in predominantly White institutions, studies on their experiences at HMIs remain limited. This study explored whether stress mediated the relationship between social comparison orientation (SCO) and two dimensions of belonging—acceptance/rejection and school membership—among Hispanic undergraduates (N = 261). Participants completed an online survey with measures of SCO, academic self-concept, belonging, school membership, and stress. Results indicated that stress fully mediated the relationship between SCO and rejection/exclusion but not acceptance/inclusion or school membership. These findings highlight the role of stress in negative social experiences among Hispanic students at HMIs. Understanding the psychological processes that shape student’s experiences in HMIs is essential for informing future research and supporting student well-being
Reproductive Natural History of the American Grass Mantid, Thesprotia graminis
Get PDFThe American grass mantis, Thesprotia graminis, is an understudied mantis native to Texas, including the campus of Texas A&M-SA. Mantids are influential members of ecosystems as a major predator of small insects. Therefore, measuring elements of their natural history, such as maternal investment, timing of oviposition, and fecundity, will add to our understanding of this specie’s ecology in grassland communities. In controlled lab conditions, we measured the mass of every instar, mantids of both sexes, and the masses of oothecas (egg cases) pre- and post-hatch. Interestingly, we found that this species lays multiple, smaller oothecas throughout the reproductive season. Using the maternal masses, ootheca masses, and offspring masses, we were able to estimate variation in reproductive investment among females and the average number of offspring produced. Additionally, we documented feeding, reproductive behavior, and the size of these species across multiple stages of development. This study represents much of the documented knowledge of the reproductive natural history of this small predator
Literature Review of Reverse Electrodialysis Technoeconomic and Environmental Analysis
Get PDFThis thesis presents a theoretical analysis of Reverse Electrodialysis (RED) for electricity generation using produced water—a high-salinity byproduct from oil and gas extraction. Drawing from over seventy peer-reviewed studies, this research explores the potential of RED as a sustainable energy solution for industrial wastewater streams. Chapter 1 outlines the environmental and economic implications of produced water. Chapter 2 reviews ion transport theories and membrane technologies. Chapter 3 evaluates past implementations of RED and related membrane systems. Chapter 4 delivers a comprehensive synthesis of current literature, modeling RED performance under realworld salinity gradients, ion compositions, and membrane constraints. Chapter 5 discusses implications for future research, scalability, antifouling technologies, and integration with energy recovery infrastructure. Emphasis is placed on membrane selectivity, thermodynamic efficiency, power density, and system sustainability. This work supports the feasibility of RED as a dual-purpose technology: reducing waste and generating renewable energy, and sets the stage for pilot-scale application in the oilfield and beyond
Fostering Latinx/e students’ familial capital in the Spanish Heritage Language Program at the University of XXXX
Get PDFIn this presentation, I show how we (director and instructors) have fostered Latinx/e students’ familial capital in the Spanish Heritage Language Program at the University of XXX. Familial capital recognizes the nurtured relationship students have with their immediate and extended family members and communities. This capital also includes the lessons students learned on caring, coping, emotional, moral, and educational consciousness from their family members. Additionally, this capital pulls from the research of Vélez-Ibáñez and Greenberg (1992) on Funds of Knowledge that points out that Mexican-Americans learn valuable lessons from their family and community that have been passed down across generations. In the SHL classroom context, familial capital focuses on creating relationships of care among students and educators. This includes building a sense of community in and outside the classroom and developing a mentorship relationship between the students and educator. This is possible by explicitly implementing Latinx/e community cultural wealth in the SHL curricula