Syracuse University
Syracuse University Research Facility and Collaborative EnvironmentNot a member yet
20376 research outputs found
Sort by
Variation on a Theme: Journey to the West and the Voice of Chinese Art
No full textThis study discusses the historical progression of adaptation of the story of Journey to the West, the 1592 novel by Wu Cheng’en and the folklore, stories, and oral tradition out of which it was born. This study primarily focuses on the characterization of the character, the Monkey King Sun Wukong, analyzing the ways and reasons Chinese artists and writers adapted this story. This thesis proposes that these adaptations served to express the respective ideas of these artists in order to promote and share them with others. This study introduces and discusses selected art forms, namely painting and carving, shadow puppetry, opera, and animation, analyzing the manner in which the Monkey King has played a role in being the voice of Chinese artists. This study posits that during various periods, the Monkey King has been used as a conversation catalyst between these art forms, creating an artistic thread of continuity and change. Ultimately, this study discusses the history of Journey to the West adaptations, focusing on the relationship of inheritance and feedback between these disparate media. At various stages in China’s history, artists and writers have played a role in domestic conversations on politics and religion as well as international conversations of diplomacy and soft power relations. The words and images used in this process have proven time and time again to be those from Journey to the West. This study outlines these connections to show how Journey to the West has been used to speak with others and the self in Chinese history
Studies Towards the Synthesis of a Simplified Forskolin Core and Phomactin A
Get PDFOxadecalins are privileged scaffolds found in various natural products that show biological activity, such as platelet aggregating factor antagonism and activation of adenyl cyclase. As such, various synthetic methods such as intramolecular substitution reactions and [4+2] cycloadditions have been developed for the construction of these important intermediates. Studies towards the synthesis of a simplified core of the labdane diterpene, forskolin, are described. Forskolin is the only known supplement to naturally raise cAMP levels, thereby displaying important pharmacological properties. The synthetic route described features a Lewis acid catalyzed dihydropyrone Diels-Alder reaction for rapidly assembly of the ABC tricycle. Modified Saegusa-Ito oxidation conditions provided access to an C7,C8 alpha-beta unsaturated enone which was later found to be prone to beta-elimination following nucleophilic attack. Introduction of the cis C6,C7 diol was achieved via a regioselective Rubottom oxidation/reduction sequence. It was later discovered that regioselective functionalization might be complicated by the stereochemistry at the ABC ring junction. Chemical elaboration the tricyclic framework led to several advanced intermediates that provided further insights into the chemistry of these oxadecalin intermediates. The platelet aggregating factor inhibitor, phomactin A, was a second synthetic target for us. Several oxadecalin intermediates were prepared using the dihydropyrone Diels-Alder reaction. Successful introduction of an allyl sidechain to a simplified oxadecalin model prompted further investigation into C6 alkylation as part of our efforts towards the synthesis of a suitable macrocyclization precursor. Several palladium catalyzed coupling conditions were also explored. Post coupling modification of an alkyne substituted oxadecalin via a zirconium assisted carboalumination/iodination sequence resulted in the formation of an unexpected enol ether
I WANT TO LOVE HOW I DO LEARN: A PHENOMENOLOGICAL STUDY OF ADULT LEARNERS WITH GENERAL LEARNING DISABILITIES AND THEIR GOALS FOR POSTSECONDARY EDUCATION
Get PDFA subset of individuals with learning disabilities is currently prevented from accessing postsecondary education; these individuals experience general learning disabilities, or GLD (Greenspan, 2017, p. 117). GLD is not a formal disability category in K-12 special education; individual students might not be eligible for academic supports and accommodations in K-12 or in postsecondary education based on this description. Individuals with GLD encounter significant barriers both accessing and finding success in postsecondary education. Qualitative research involving adults with GLD is lacking in available literature, as is theoretical analysis of their experiences. My research is a phenomenological study based on the experiences of ten adult learners with GLD who had a strong desire to participate in postsecondary education but often found themselves thwarted due to admissions requirements. The ten participants, each in a set of three interviews, were asked to reflect upon their understanding of themselves as learners in formal K-12 settings, their personal areas of mastery in adulthood, and their hopes to change as learners by participating in postsecondary education. Analysis of the interview transcripts resulted the following insights: the participants were highly aware of their difficulties in formal education settings (self-aware outsiders); the participants were actively engaged in pursuing valuable information outside of school (directors of their own learning); and the participants hoped to gain confidence as well as skills by participating in postsecondary education (claiming learner identity). Five of the ten participants identified as people of color. Tenets of Disability Critical Race Theory (DisCrit) and other concepts impacting multiply marginalized students (hope theory, critical hope, and intellectual activism) informed the analysis of the participants’ experiences
THE LIVED EXPERIENCES OF FEMALE UNDERGRADUATES AT A PRIVATE UNIVERSITY REGARDING FOODWAYS AND WELLNESS: AN INTERPRETIVE PHENOMENOLOGICAL ANALYSIS
Get PDFThis dissertation study outlines the topic of food and wellness for female undergraduate college students at a private university in the northeastern US. An extensive literature review on the topic of human interactions with food and wellness as well as the potential impact on female undergraduate students is provided, followed by a proposed qualitative study. This dissertation describes the qualitative method of Interpretive Phenomenological Analysis (IPA) using semi-structured interviews and member-checking meetings to learn about the experiences of eight female undergraduate students at a private university’s experiences with food, eating, sourcing, cooking, and sharing and how these behaviors impact and are impacted by their overall wellness. Five themes were pulled from participants’ experiences including (1) Environmental Context and Foodways, (2) Physical Impacts of Foodways Aside from Weight, (3) Connection and Hinderance, (4) Accessibility and Agency, and (5) Oppression and Resistance. These themes and their subthemes as well as similar and differing experiences between participants are shared. Charts are included throughout the Findings Chapter to better illustrate connections between themes. Implications for addressing gaps in clinical counseling work and counselor training will be discussed. Definitions of key terms such as wellness, food insecurity, and food systems are provided and discussed at length, and interview questions are provided in the appendices
EFFECTS OF SELF-DETERMINATION THEORY (SDT)-BASED INSTRUCTIONAL STRATEGIES ON LEARNER ENGAGEMENT IN ASYNCHRONOUS ONLINE DISCUSSIONS (AODS)
Get PDFAbstract Learner engagement is a malleable state influenced by various contextual factors. Prior research has identified several elements that affect engagement, including learning styles, course perception, computer self-efficacy, and a sense of presence. Furthermore, it is imperative to consider both learner engagement dimensions and environmental affordances when investigating learner engagement. This study, grounded in Self-Determination Theory (SDT), investigates the effects of SDT-based instructional strategies on learner engagement in asynchronous online discussions (AODs). The primary aim is to examine how these strategies, designed to support learners perceived psychological needs of autonomy, relatedness, and competence, affect behavioral, cognitive, and emotional dimensions of engagement. Utilizing a quasi-experimental, sequential mixed-methods design, the study was conducted over 15 weeks with 32 learners in a graduate-level course. It involved a comparative condition (non-SDT AOD) and an experimental condition (SDT-based AOD), aimed to address the following research questions: 1) How do SDT-based instructional strategies impact learner engagement in AODs? 2) How do learners perceived psychological needs of autonomy, relatedness, and competence relate to their engagement dimensions in SDT-based AODs? 3) How do learners perceive the influence of SDT-based instructional strategies on their engagement? The study began with a quantitative phase, followed by qualitative data collection and analysis. Multiple sources of data were collected, including self-reported survey, semi- structured interview, and discussion discourse. Findings revealed a significant increase in learners perceived autonomy for SDT-based AODs. While perceived relatedness and competence did not show significant differences between the two conditions, these factors still played a crucial role in enhancing learners’ actual engagement as evidenced by the discussion discourse analysis. Social network analysis (SNA) showed SDT-based AODs fostered stronger interactions and a more inclusive environment. Content analysis with discussion discourse indicated higher cognitive and emotional engagement in SDT-based AODs, corroborated by interview data highlighting enhanced continuous engagement, critical thinking, and a stronger sense of community. This study significantly advances the theoretical understanding of online learning by examining AOD instructional strategies through the lens of SDT. It provides empirical insights into effective instructional designs and uniquely treats learner engagement as a multi- dimensional construct, offering a comprehensive look at how SDT-based instructional strategies impact behavioral, cognitive, and emotional engagement. Using a mixed-methods approach, the research captures the dynamic nature of engagement more effectively than self-reports alone. Grounded in SDT, the study suggests instructional strategies that leverage motivational principles to enhance interaction and higher-order cognitive processing in AODs, underscoring the importance of designing online learning environments that fulfill learners\u27 psychological needs. It also explores the independent effects of satisfying autonomy, competence, and relatedness on different engagement dimensions, offering a nuanced understanding for more targeted strategies. These contributions enhance the implementation of effective online instructional strategies, leading to improved learner engagement and academic outcomes
Enhancing Graph Neural Networks by Editing Graphs
Get PDFGraphs are pervasive in both the natural world and various domains of science and engineering. Numerous advanced classifiers, such as Graph Neural Networks (GNNs), have been developed to perform node classification on these graphs. However, as graphs become denser with an increasing number of edges, GNNs often suffer from suboptimal generalization performance due to the presence of task-irrelevant connections. These redundant connections can introduce noise, consume excessive computational resources, and degrade performance. Identifying and preserving critical connections in large-scale graphs, while pruning unnecessary ones, is crucial for enhancing the efficiency and accuracy of GNNs in node classification, particularly for GCNs. As a result, the first challenge is determining whether it is possible to identify skeletal substructures within graphs—key subsets of connections—that can preserve the predictive performance of node classification. This involves either maintaining accuracy when using these critical subgraphs instead of the entire graph or training classifiers on these subgraphs with minimal performance degradation. To address the challenge, we propose the Sparsified Graph Convolutional Network (SGCN), a neural network-based graph sparsifier. SGCN effectively reduces graph density by pruning certain edges while still maintaining comparable performance in node classification tasks. To further improve the efficiency of GNN with even more sparse graph, we introduce the Graph Ultra-sparsifier, a semi-supervised graph sparsification method that incorporates dynamically-updated regularization terms derived from graph convolution. This approach preserves the properties of graph filters, enabling the generation of sparser graphs that maintain the performance of GCN models when used as input. To further explore the significance of key subsets of connections, we propose a sparse adversarial attack framework, AdverSparse. This framework demonstrates how the removal of just a few critical connections can severely disrupt the spatial dependencies learned by spatial-temporal models, leading to issues such as increased prediction errors. While graph sparsification effectively addresses the space and time consumption issues of GCNs while maintaining their performance, existing graph editing methods are specifically designed for the low-pass filter structure of GCNs, which inherently favor homophilic graphs. However, real-world graphs often exhibit varying levels of homophily. To further enhance the capability of GCNs on graphs with diverse homophily ratios, it is essential to go beyond merely removing edges. Instead, we propose dynamically assigning weights to both existing edges and hidden edges (i.e., edges from complement graphs). This introduces a second challenge: dynamically editing graphs. The goal is to modify graphs by incorporating edges from both the original graph and its complement in a way that allows them to serve as adaptive graph filters. Such filters should dynamically adjust their behavior based on the homophily ratio of the graph, improving node classification performance. To address the second challenge, we introduce Complement Graph Convolution (CGC), a method designed to overcome the limitations of GCNs and enhance the performance of GNNs that use GCNs as basic modules for node classification tasks across graphs with varying homophily levels. CGC incorporates a trainable, adaptive frequency-response filter that utilizes both an original graph and its complement graph. Spectral analysis reveals that this filter can be expressed as a Bernstein polynomial approximation, allowing it to dynamically adjust the importance of different frequency components for spectral convolution, effectively accommodating diverse homophily and heterophily ratios. Furthermore, we demonstrate that CGC mitigates the over-smoothing problem commonly associated with GCNs. The overall objective of this thesis is to explore methods for enhancing the efficiency and capability of GNNs, particularly the GCN family, by editing graph structures to affect the properties of graph filters
Examining the Extent of Exculsionary Zoning Practices in Northern Lackawanna County, Pennsylvania using the Surrick Fair Share Test
Get PDF
