93887 research outputs found
Sort by
Comparing human text classification performance and explainability with large language and machine learning models using eye-tracking
Get PDFTo understand the alignment between reasonings of humans and artificial intelligence (AI) models, this empirical study compared the human text classification performance and explainability with a traditional machine learning (ML) model and large language model (LLM). A domain-specific noisy textual dataset of 204 injury narratives had to be classified into 6 cause-of-injury codes. The narratives varied in terms of complexity and ease of categorization based on the distinctive nature of cause-of-injury code. The user study involved 51 participants whose eye-tracking data was recorded while they performed the text classification task. While the ML model was trained on 120,000 pre-labelled injury narratives, LLM and humans did not receive any specialized training. The explainability of different approaches was compared based on the top words they used for making classification decision. These words were identified using eye-tracking for humans, explainable AI approach LIME for ML model, and prompts for LLM. The classification performance of ML model was observed to be relatively better than zero-shot LLM and non-expert humans, overall, and particularly for narratives with high complexity and difficult categorization. The top-3 predictive words used by ML and LLM for classification agreed with humans to a greater extent as compared to later predictive words
Factors Related to Study Progress Among First-Year Agriculture Students
Get PDFThe first study year at university predicts the progress and quality of later studies. The aim of our study was to explore factors that affect first-year agriculture students. In the end of their first year, 49 students answered a questionnaire measuring self-efficacy, approaches to learning, and study-related burnout. They also reported the factors that enhanced or impeded their studies. According to their approaches to learning, students were clustered into three profiles. One of these represented successful students with an organized approach, strong self-efficacy and little burnout, and another a more unorganized group. The third group had a dissonant profile and suffered from the highest burnout levels. The enhancing factors most often mentioned were peer support and regular assignments. These were recognized by the organized group. The most common impeding factors were activities outside of studies, recognized by the unorganized group, and high workload, recognized mostly by the dissonant group. Pedagogical implications are discussed
“It’s like listening to an audiobook”: University of Botswana Students’ Lived Experiences and Views about Online Learning during the COVID-19 Pandemic
Get PDFThe abrupt shift to online teaching and learning during the COVID-19 pandemic put a strain on the education sector the world over. Using the social constructivist theory, this study postulates that faceto- face teaching and learning does not seamlessly translate to online instruction both in terms of pedagogical practice and learner experiences. This study explores students’ experiences with the quality of online learning during the pandemic. Data were collected through focus group discussions with undergraduate students across seven faculties. A thematic analysis of the responses reveals that participants mostly reported negative experiences with online learning arising from factors such as lack of compatible digital devices and conducive virtual class learning spaces. This study provides rich data that contributes to an understanding of students’ experiences with online teaching and learning during the pandemic and thus, provides insights into how lecturers’ online pedagogical practices influence students’ perceptions on the quality of online learning
The Effects of Math Anxiety on the Performance of Undergraduate Business Majors: Using Self-Efficacy as a Mediator
Get PDFThis research contributes to the body of knowledge regarding mathematics anxiety, self-efficacy, and performance in mathematics. Specifically, this study analyzed these constructs as they pertain to undergraduate business students enrolled in entry-level, prerequisite mathematics courses. Information was collected via surveys utilizing the Mathematics Self-Efficacy and Anxiety Questionnaire (MSEAQ). Results based on regression modeling were consistent with prior research involving the relationship between mathematics anxiety and performance, with self-efficacy serving as a mediator. Data indicated an inverse relationship between math anxiety and math self-efficacy, an inverse relationship between math anxiety and students’ expected grade, differences in math self efficacy by business major, and partial mediation support for math self-efficacy on the inverse relationship between math anxiety and expected grade. Discussion extends to instructional strategies for mathematics and business educators alike that support self-efficacy and alleviate mathematics anxiety for business students in the infancy of their program
Developing nursing education in a low-income country
No full textBackground Nursing schools in low-income countries face challenges to effectively educate and retain future nurses in rural settings where healthcare needs are more prominent. It is important to understand the experiences of nursing students in rural settings to better prepare them to meet the complex needs of the community. Little is known about nursing students\u27 experiences in rural Uganda. The purpose of this paper is to explore nursing students\u27 experiences of attending school in a rural setting in Uganda. Methods A qualitative descriptive design was used to conduct this study. Twelve nursing students from a nursing school in rural Uganda participated in semi-structured interviews. Data were analyzed using an inductive content analysis. Results Five themes were developed from the data to describe the participants\u27 experiences attending nursing school in rural Uganda. The five themes consisted of: 1) External Factors Affect Nursing Students\u27 Experiences, 2) Physical Environment Features Influence Nursing Students\u27 Experiences, 3) Teaching and Mentoring Capabilities Shape Nursing Students\u27 Experiences, 4) Technologies Impact Nursing Students\u27 Experiences, and 5) Clinical Environment Considerations Structure Nursing Students\u27 Experiences. Discussion/implications Nursing students identified strengths and weaknesses of the learning environment which shaped their experiences of attending nursing school in rural Uganda. The findings of this study also highlight the need for more teaching resources, simulation technology, and on-site facilities to enhance the learning environment. Implications for improving nursing education in rural settings in Uganda and other low-income countries include recruiting and retaining more qualified faculty, offering professional development opportunities to faculty, and developing infrastructure to support quality education and training
Crafting the Wings of Tomorrow\u27s Leaders
Get PDFSOAR is a social impact initiative I founded in partnership with Big Brothers/Big Sisters Lafayette. We provide novel leadership and character development resources for local Lafayette youth. The motto—engage, uplift, and achieve—speaks to my aims for inspiring the next generation of leaders, within Purdue members and local youth. This reflection essay is an opportunity to express the aims of SOAR, the intentionality behind this initiative, and how I hope to contribute to the development of youth in Lafayette and beyond. Mr. Brennan’s visit is a milestone for SOAR, and having the opportunity to converse with a renowned youth development leader was inspirational. Sharing this endeavor’s roots with Mr. Brennan, learning sustainability measures, and discovering ways to create transformative impact is a chief aim of mine. I also greatly value the chance to reflect on how SOAR has changed my outlook on leadership and service. SOAR has given me a platform to help fundamentally alter how youth view themselves and their capabilities. My greatest goal is providing youth avenues to learn leadership competencies, emotional regulation, and character development. Our programming is founded on empirical research, allowing SOAR to ensure strong interconnectivity between research and programming. Building SOAR into a robust organization has been a significant undertaking. Connecting and collaborating innovatively with peers has been deeply rewarding. Throughout the 2023–2024 academic year, SOAR has tripled in membership, conducted 20+ hours in volunteer efforts, and made 6 visits to Lafayette Mentors. I look forward to deepening the relationship with Big Brothers, Big Sisters. Through the Lead Forward fellowship, I have learned about the power of transformational leadership. As I step into the role of service leadership, I am eager to become more knowledgeable of providing empowerment and transformation to the youth—but also to my peers in SOAR
Promoting Cultural Belonging through the Arts: Community Partner Snapshot of Latino Art Midwest
Get PDFKayla Vasilko just completed her Master’s degree in communication with a focus on semiotics and the impact of higher education on positive social change in the Purdue Graduate School and the PNW Department of Communication and Creative Arts. While earning her degree, she had the opportunity to complete a service-learning project to celebrate Spanish culture through the arts and work the community partner Latino Art Midwest. Latino Art Midwest works to promote understanding of the role of arts and creative expression among Midwestern Latina/o communities. Students can explore their site, “view the art, and interpret the unique qualities of Latina/o artistic expression in the Midwest to discover how the arts attest to all kinds of transnational experiences—from migration, to bi-national identities, to the social networks that Latina/o artists are a part of—that ultimately illuminate how aesthetic cultural production has played a role in forming strong and vibrant communities that make the Midwest global” (Latino Art Midwest, 2024). Research, reflection, university art loans, and other engagement with Latino Art Midwest create important opportunities to combat cultural discrimination and increase representation
Identification of Conserved Eukaryotic Gene Products in Novel Mycobacteriophage
Get PDFBacteriophage, a type of virus that infects bacteria, was discovered 100 years ago, initially isolated with a plaque assay in the 1940s. A mycobacteriophage is a subtype of bacteriophage that specifically infects mycobacteria. Since the introduction of DNA technologies and efforts to identify a plethora of bacteriophages, over 10,000 bacteriophages have been identified with 1,500 sequenced. Bacteriophage discovery has aided in genetic editing techniques, the development of synthetic biology tools, and disease treatment. However, to be able to expand on these capabilities, there is a need to identify more bacteriophages and sequence the genomes of these viruses, as they outnumber bacteria 10 to 1. The mycobacteriophage PurduePete was discovered in 2022 and annotated using the gene sequence annotation programs DNA Master, PhagesDB BLAST, and NCBI BLAST to discover the functions of its genome. Several encoded gene products were discovered in PurduePete that are part of larger, biologically significant, protein families present in eukaryotes, prokaryotes, and other bacteriophage suggesting possible conservation and homology in PurduePete. This includes the functions of helix-turn-helix DNA binding protein in the base pair (bp) region 17644 to 17947, histidine triad protein (HIT) in the bp region 6280 to 6858, and a RusA-like Holliday junction resolvase in the bp region 121254–121709 bp of PurduePete. Developing a greater understanding of shared gene products between bacteriophages and eukaryotes will be integral in creating bacteriophage-based technologies that support already existing eukaryotic functions. The aforementioned proteins are all therapeutically significant, spanning areas of oncology and cardiology. These protein functions present in bacteriophages could be scaled and utilized in the development of novel therapeutics and provide a framework to investigate their broader functions
Predictive Maintenance Analysis of Turbofan Engine Sensor Data
Get PDFPredictive maintenance in aviation and aerospace applications is among the most explored problems in machine learning (ML) and artificial intelligence (AI), and datasets such as NASA’s C-MAPPS turbofan engine degradation simulation data have proven invaluable, helping researchers explore numerous questions on engine performance, maintenance, and failure. The purpose of this study was to extend the current research on predicting the remaining useful life (RUL) of engines and their risk classification. Starting with simple yet under-investigated nonlinear survival and random forest models, the analysis implemented eXtreme Gradient Boosting (XGBoost) and long short-term memory (LSTM) from TensorFlow’s Keras library. For both regression and classification, valuable insights were gained from principal component analysis (PCA), lifelines survival analysis, k-nearest neighbors (KNN) algorithms, and random forests. The regression results obtained from XGBoost and LSTM architectures were comparable to those reported in related studies. The XGBoost regressor outperformed LSTM, achieving a root mean squared error (RMSE) as low as 21.29 and an R-squared score of 0.74. The XGBoost classifier yielded accuracy scores as high as 90%, matching the performance of LSTM and random forest classifiers
Effect of Developmental Manganese Exposure on the Locomotor Behavior of Zebrafish Larvae
Get PDFManganese, an essential trace mineral, plays crucial roles in various physiological processes within living organisms. However, excessive exposure to manganese can lead to toxicity, particularly impacting the central nervous system. This study aimed to assess the impact of manganese exposure on locomotor behavior during early development using zebrafish (Danio rerio) as a model organism. Zebrafish embryos at 1–2 hours post fertilization (hpf) were exposed to five different concentrations of manganese chloride (0, 0.01, 0.05, 0.1, or 0.5 mM) until 120 hpf. Locomotor behavior was assessed using the visual motor response assay, measuring parameters such as total distance moved, velocity, and time spent moving. Additional assessments included spontaneous movement at 24 hpf, heart rate at 48 hpf, and survival and hatching rates from 24 to 120 hpf. Results indicated significant alterations in locomotor behavior, with decreased total distance moved and velocity observed in higher concentration groups during dark phases. Interestingly, the lowest concentration group exhibited hyperactivity during light phases, suggesting a dose-dependent effect on behavior. Spontaneous movement significantly increased at 0.05 mM concentration, while heart rate decreased at the highest concentration. Hatching rates were elevated in the two highest concentration groups at 24 hpf. These findings underscore the association between manganese exposure and locomotor behavioral changes during early development. Moreover, the differential effects observed across concentration groups suggest varying mechanisms underlying manganese toxicity. Zebrafish serve as a valuable model for investigating developmental neurotoxicity, providing insights relevant to human health outcomes