ASU Electronic Theses and Dissertations
This collection includes most of the ASU Theses and Dissertations from 2011 to present. ASU Theses and Dissertations are available in downloadable PDF format; however, a small percentage of items are under embargo. Information about the dissertations/theses includes degree information, committee members, an abstract, supporting data or media.
In addition to the electronic theses found in the ASU Digital Repository, ASU Theses and Dissertations can be found in the ASU Library Catalog.
Dissertations and Theses granted by Arizona State University are archived and made available through a joint effort of the ASU Graduate College and the ASU Libraries. For more information or questions about this collection contact or visit the Digital Repository ETD Library Guide or contact the ASU Graduate College at gradformat@asu.edu.
Displaying 1 - 10 of 327
Filtering by
- All Subjects: Machine Learning
- All Subjects: Gender
Description
Sparsity has become an important modeling tool in areas such as genetics, signal and audio processing, medical image processing, etc. Via the penalization of l-1 norm based regularization, the structured sparse learning algorithms can produce highly accurate models while imposing various predefined structures on the data, such as feature groups or graphs. In this thesis, I first propose to solve a sparse learning model with a general group structure, where the predefined groups may overlap with each other. Then, I present three real world applications which can benefit from the group structured sparse learning technique. In the first application, I study the Alzheimer's Disease diagnosis problem using multi-modality neuroimaging data. In this dataset, not every subject has all data sources available, exhibiting an unique and challenging block-wise missing pattern. In the second application, I study the automatic annotation and retrieval of fruit-fly gene expression pattern images. Combined with the spatial information, sparse learning techniques can be used to construct effective representation of the expression images. In the third application, I present a new computational approach to annotate developmental stage for Drosophila embryos in the gene expression images. In addition, it provides a stage score that enables one to more finely annotate each embryo so that they are divided into early and late periods of development within standard stage demarcations. Stage scores help us to illuminate global gene activities and changes much better, and more refined stage annotations improve our ability to better interpret results when expression pattern matches are discovered between genes.
ContributorsYuan, Lei (Author) / Ye, Jieping (Thesis advisor) / Wang, Yalin (Committee member) / Xue, Guoliang (Committee member) / Kumar, Sudhir (Committee member) / Arizona State University (Publisher)
Created2013
Description
This work presents two complementary studies that propose heuristic methods to capture characteristics of data using the ensemble learning method of random forest. The first study is motivated by the problem in education of determining teacher effectiveness in student achievement. Value-added models (VAMs), constructed as linear mixed models, use students’ test scores as outcome variables and teachers’ contributions as random effects to ascribe changes in student performance to the teachers who have taught them. The VAMs teacher score is the empirical best linear unbiased predictor (EBLUP). This approach is limited by the adequacy of the assumed model specification with respect to the unknown underlying model. In that regard, this study proposes alternative ways to rank teacher effects that are not dependent on a given model by introducing two variable importance measures (VIMs), the node-proportion and the covariate-proportion. These VIMs are novel because they take into account the final configuration of the terminal nodes in the constitutive trees in a random forest. In a simulation study, under a variety of conditions, true rankings of teacher effects are compared with estimated rankings obtained using three sources: the newly proposed VIMs, existing VIMs, and EBLUPs from the assumed linear model specification. The newly proposed VIMs outperform all others in various scenarios where the model was misspecified. The second study develops two novel interaction measures. These measures could be used within but are not restricted to the VAM framework. The distribution-based measure is constructed to identify interactions in a general setting where a model specification is not assumed in advance. In turn, the mean-based measure is built to estimate interactions when the model specification is assumed to be linear. Both measures are unique in their construction; they take into account not only the outcome values, but also the internal structure of the trees in a random forest. In a separate simulation study, under a variety of conditions, the proposed measures are found to identify and estimate second-order interactions.
ContributorsValdivia, Arturo (Author) / Eubank, Randall (Thesis advisor) / Young, Dennis (Committee member) / Reiser, Mark R. (Committee member) / Kao, Ming-Hung (Committee member) / Broatch, Jennifer (Committee member) / Arizona State University (Publisher)
Created2013
Description
This dissertation integrates humanities with social science methodologies within a critical framework, seeking to explore the relationship between the neoliberal restructuring and the intersection of gender, class and heteronormativity in contemporary China. In this project, neoliberalism is conceptualized as an art of governance centering on the intersection of race, gender, class and sexuality to create market subjects and sustain market competition. Focusing on China's recent socio-economic and cultural upheavals, this dissertation tries to address these questions: 1. How have class inequalities, binaristic gender and heteronormative discourses been employed intersectionally by the Chinese state to facilitate China's social transformation? 2. How has this process been justified and consolidated through the intersection of gender, class, sexuality and race? 3. How do the marginalized groups respond to these material and cultural practices? Building on the discursive analysis of China's televised 60th anniversary ceremony and If You Are the One, a popular Chinese reality show, as well as the data from the interview, focus group and participant observation of more than 100 informants, it is found that the intersection of gender, class and heteronormativity is central to China's neoliberal transition. A group of flexible and cheap laborers have been disarticulated and rearticulated from the population as the voluntary servitude to China's marketization and re-integration with the global economy. New controlling images, such as the bourgeois nucleus family, are created to legitimize this process. However, these disparate material and discursive practices have entailed contradictions and conflicts within the intersectional biopolitical system, and created contingent spaces of ungovernability for the marginalized groups. Building on these discursive analyses and empirical data, I reconceptualize intersectionality as a multi-dimensional-and-directional network to regulate and manage power for social organization and regulation, which grounds the biopolitical basics for the neoliberal economy. Thus I argue that we need to engage with the dynamics between the intersectional biopolitical structure and people's emerging experiences to construct a grounded utopia alternative to the neoliberal dominance for substantive social changes.
ContributorsZhang, Charlie Yi (Author) / Quan, H. L. T. (Thesis advisor) / Fonow, Mary Margaret (Thesis advisor) / Martinez, Jacqueline M. (Committee member) / Lee, Charles T. (Committee member) / Arizona State University (Publisher)
Created2013
Description
Gender and sex are often conflated. Our laws, policies, and even science establish sex and gender as intrinsically linked and dimorphic in nature. This dissertation examines the relationship between sex and gender and the repercussions of this linked dimorphism in the realms of law, politics, and science. Chapter One identifies the legal climate for changing one's sexual identity post-surgical reassignment. It pays particular attention to the ability of postsurgical transsexuals to marry in their acquired sex. Chapter Two considers the process for identifying the sex of athletes for the purposes of participation in sex-segregated athletic events, specifically the role of testing and standards for categorization. Chapter Three explores the process of identifying and assigning the sex of intersex children. Chapter Four examines the process of prenatal sex selection and its ethical implications. Chapter Four also offers an anticipatory governance framework to address these implications.
ContributorsParsi, John (Author) / Crittenden, Jack (Thesis advisor) / Guston, David H. (Committee member) / Marchant, Gary (Committee member) / Arizona State University (Publisher)
Created2013
Description
In the latter half of the nineteenth century, colleges and universities transformed their thinking of the body as they institutionalized physical education, recreational activities, and especially physical exercise. In this study, I examine the historical discourse on physical exercise and training during this period. I employ the theoretical and methodological practices of Michel Foucault's archeological and genealogical work to write a "history of the present." I challenge the essential narrative of physical fitness on college and university campuses. I also discuss nineteenth century notions of ethics and masculinity as a way of understanding twenty-first century ethics and masculinity. Ultimately, I use the historical discourse to argue that institutionalization of recreation and fitness centers and activities have less to do with health and well-being and more to do with disciplining bodies and controlling individuals.
ContributorsWells, Timothy (Author) / Carlson, David L. (Thesis advisor) / Sandlin, Jennifer (Committee member) / Margolis, Eric (Committee member) / Arizona State University (Publisher)
Created2012
Description
Arnold Schoenberg's 1908-09 song cycle, Das Buch der hängenden Gärten [The Book of the Hanging Gardens], opus 15, represents one of his most decisive early steps into the realm of musical modernism. In the midst of personal and artistic crises, Schoenberg set texts by Stefan George in a style he called "pantonality," and described his composition as radically new. Though stylistically progressive, however, Schoenberg's musical achievement had certain ideologically conservative roots: the composer numbered among turn-of-the-century Viennese artists and thinkers whose opposition to the conventional and the popular--in favor of artistic autonomy and creativity--concealed a reactionary misogyny. A critical reading of Hanging Gardens through the lens of gender reveals that Schoenberg, like many of his contemporaries, incorporated strong frauenfeindlich [anti-women] elements into his work, through his modernist account of artistic creativity, his choice of texts, and his musical settings. Although elements of Hanging Gardens' atonal music suggest that Schoenberg valued gendered-feminine principles in his compositional style, a closer analysis of the work's musical language shows an intact masculinist hegemony. Through his deployment of uncanny tonal reminiscences, underlying tonal gestures, and closed forms in Hanging Gardens, Schoenberg ensures that the feminine-associated "excesses" of atonality remain under masculine control. This study draws upon the critical musicology of Susan McClary while arguing that Schoenberg's music is socially contingent, affected by the gender biases of his social and literary milieux. It addresses likely influences on Schoenberg's worldview including the philosophy of Otto Weininger, Freudian psychoanalysis, and a complex web of personal relationships. Finally, this analysis highlights the relevance of Schoenberg's world and its constructions of gender to modern performance practice, and argues that performers must consider interrelated historical, textual, and musical factors when interpreting Hanging Gardens in new contexts.
ContributorsGinger, Kerry Anne (Author) / FitzPatrick, Carole (Thesis advisor) / Dreyfoos, Dale (Committee member) / Mook, Richard (Committee member) / Norton, Kay (Committee member) / Ryan, Russell (Committee member) / Arizona State University (Publisher)
Created2012
Description
Image understanding has been playing an increasingly crucial role in vision applications. Sparse models form an important component in image understanding, since the statistics of natural images reveal the presence of sparse structure. Sparse methods lead to parsimonious models, in addition to being efficient for large scale learning. In sparse modeling, data is represented as a sparse linear combination of atoms from a "dictionary" matrix. This dissertation focuses on understanding different aspects of sparse learning, thereby enhancing the use of sparse methods by incorporating tools from machine learning. With the growing need to adapt models for large scale data, it is important to design dictionaries that can model the entire data space and not just the samples considered. By exploiting the relation of dictionary learning to 1-D subspace clustering, a multilevel dictionary learning algorithm is developed, and it is shown to outperform conventional sparse models in compressed recovery, and image denoising. Theoretical aspects of learning such as algorithmic stability and generalization are considered, and ensemble learning is incorporated for effective large scale learning. In addition to building strategies for efficiently implementing 1-D subspace clustering, a discriminative clustering approach is designed to estimate the unknown mixing process in blind source separation. By exploiting the non-linear relation between the image descriptors, and allowing the use of multiple features, sparse methods can be made more effective in recognition problems. The idea of multiple kernel sparse representations is developed, and algorithms for learning dictionaries in the feature space are presented. Using object recognition experiments on standard datasets it is shown that the proposed approaches outperform other sparse coding-based recognition frameworks. Furthermore, a segmentation technique based on multiple kernel sparse representations is developed, and successfully applied for automated brain tumor identification. Using sparse codes to define the relation between data samples can lead to a more robust graph embedding for unsupervised clustering. By performing discriminative embedding using sparse coding-based graphs, an algorithm for measuring the glomerular number in kidney MRI images is developed. Finally, approaches to build dictionaries for local sparse coding of image descriptors are presented, and applied to object recognition and image retrieval.
ContributorsJayaraman Thiagarajan, Jayaraman (Author) / Spanias, Andreas (Thesis advisor) / Frakes, David (Committee member) / Tepedelenlioğlu, Cihan (Committee member) / Turaga, Pavan (Committee member) / Arizona State University (Publisher)
Created2013
Description
Although aggression is sometimes thought to be maladaptive, evolutionary theories of resource control and dominance posit that aggression may be used to gain and maintain high social prominence within the peer group. The success of using aggression to increase social prominence may depend on the form of aggression used (relational versus physical), the gender of the aggressor, and the prominence of the victim. Thus, the current study examined the associations between aggression and victimization and social prominence. In addition, the current study extended previous research by examining multiple forms of aggression and victimization and conceptualizing and measuring social prominence using social network analysis. Participants were 339 6th grade students from ethnically diverse backgrounds (50.4% girls). Participants completed a peer nomination measure assessing relational and physical aggression and victimization. They also nominated friends within their grade, which were used to calculate three indices of social prominence, using social network analysis. As expected, results indicated that relational aggression was associated with higher social prominence, particularly for girls, whereas physical aggression was less robustly associated with social prominence. Results for victimization were less clear, but suggested that, for girls, those at mid-levels of social prominence were most highly victimized. For boys, results indicated that those both high and low in prominence were most highly relationally victimized, and those at mid-levels of prominence were most highly physically victimized. These findings help inform intervention work focused on decreasing overall levels of aggressive behavior.
ContributorsAndrews, Naomi C. Z (Author) / Hanish, Laura D. (Thesis advisor) / Martin, Carol Lynn (Committee member) / Updegraff, Kimberly A (Committee member) / Arizona State University (Publisher)
Created2013
Description
This is a hermeneutic study on experiences being gifted, teaching gifted students and/or raising gifted children. This study focuses on how our horizon, which is a result of our past experiences, has an impact on how we make sense of our world and influences our attitudes and actions. As became clear during the conduct of the research, gender was the dominant characteristic of the horizon and unconscious hermeneutic processes these women used to make sense of their experiences. Gender, it became clear also impacted their self-understanding of who they were, what were their possibilities in life, and the decisions they now make as parents and teachers. For this study the researcher interviewed twelve teachers and parents from two different districts who are involved in gifted programs. Some of them had children involved in gifted classes, some were in gifted programs as a child, some worked in gifted programs as an adult and some were a combination of the three. Data consisted of twelve original interviews. Four of the original twelve were selected and each was interviewed a second time. Data from both interviews was analyzed hermeneutically. Included in the study are each participant's horizon and a topical analysis of the interviews. In addition, a thematic analysis is included which ties each interview to themes and cultural norms.
ContributorsHarrison, Erin Kirkpatrick (Author) / Blumenfeld-Jones, Donald (Thesis advisor) / Rutowski, Kathleen (Committee member) / Carlson, David Lee (Committee member) / Arizona State University (Publisher)
Created2012
Description
With the increase in computing power and availability of data, there has never been a greater need to understand data and make decisions from it. Traditional statistical techniques may not be adequate to handle the size of today's data or the complexities of the information hidden within the data. Thus knowledge discovery by machine learning techniques is necessary if we want to better understand information from data. In this dissertation, we explore the topics of asymmetric loss and asymmetric data in machine learning and propose new algorithms as solutions to some of the problems in these topics. We also studied variable selection of matched data sets and proposed a solution when there is non-linearity in the matched data. The research is divided into three parts. The first part addresses the problem of asymmetric loss. A proposed asymmetric support vector machine (aSVM) is used to predict specific classes with high accuracy. aSVM was shown to produce higher precision than a regular SVM. The second part addresses asymmetric data sets where variables are only predictive for a subset of the predictor classes. Asymmetric Random Forest (ARF) was proposed to detect these kinds of variables. The third part explores variable selection for matched data sets. Matched Random Forest (MRF) was proposed to find variables that are able to distinguish case and control without the restrictions that exists in linear models. MRF detects variables that are able to distinguish case and control even in the presence of interaction and qualitative variables.
ContributorsKoh, Derek (Author) / Runger, George C. (Thesis advisor) / Wu, Tong (Committee member) / Pan, Rong (Committee member) / Cesta, John (Committee member) / Arizona State University (Publisher)
Created2013