We've Moved!

Visit SDSU’s new digital collections website at https://digitalcollections.sdsu.edu

Collection Description

Collection of student theses and dissertations from as early as 1939, but mainly from 2010 to present.

Back to top

Pages

Who is the speaker? A learning based audio-visual approach
This thesis presents a learning-based audio-visual approach for detecting active speakers in videos of group meetings. The video inputs often include an unknown number of humans with low-resolution facial regions, which pose great challenges to traditional vision methods. The key idea of this effort is to employ both visual inputs and audio signals to boost detection accuracy. The major contributions are threefold. Firstly, a learning-based object detection method is developed to localize human faces in videos and classify them as either speaker or non-speaker according to their facial features. Secondly, the synchronized audio signals are processed to estimate the existence of the speaker over time. Thirdly, a comprehensive fusion strategy for speaker detection is introduced to combine the results of visual analysis and the results of audio analysis. The proposed methods were evaluated on a newly collected video dataset that includes hundreds of YouTube videos. Experimental results showed that the audio-visual method achieved substantial improvements over visual methods., San Diego State University
Who is there and what are they doing? An agile and computationally efficient framework for genome discovery and annotation from metagenomic big data
Microbes are more abundant than any other biological guild, and in any environment it is important to understand which organisms are present, what they are doing, and how they are doing it. In many environments a majority of the microbial community members cannot be cultured. Metagenomics is a powerful tool to directly probe uncultured genomes and understand the diversity of microbial communities using only their DNA sequences. Analyzing the taxonomic and functional profiles present in a microbial community from unannotated shotgun sequencing reads is one of the goals in metagenomics, with extremely valuable applications in biological research such as medicine, biofuels, and ecology. Currently available tools do not scale well with increasing data volumes, which is important because both the number and lengths of the reads produced by sequencing platforms keeps increasing. This thesis integrates four agile and computationally efficient methods that I have developed (FOCUS, FOCUS2, Scaffold builder, and SUPER-FOCUS) to recover, scaffold, and annotate genomes from metagenomes. The framework was tested in over 500 human and ocean samples totaling over 6TB of data, and over six thousand genomes were recovered. Each computational method presented in this dissertation opens new horizons for the future of metagenomic data analyses independently of query and database size.
Who we are when we say we are : the politics of slam poetry
Includes bibliographical references (pages 53-54)., This study examines spoken word poetry and poetry slam competitions in regards to utopianism, identity politics, and actions of healing. My research methodologies include performance art theory and emotional theories—all through an intertwined feminist lens. This study will explore how possible intersecting identities function in the poetry slam world and how those intersecting identities create differences in definitions of "utopia." Further, the role of "emotion" in the world of slam will also be examined, specifically from the perspective of interviewees/spoken word performers. The purpose of this study is to explore how identity influences success in competitive poetry and success in actions of healing for marginalized voices. A total of six previous members of the San Diego Slam Team (ranging from 2008-2012) were interviewed for this research. All subjects involved in this study play a role in my analyses of how competitors in slam poetry employ their own emotions, experiences, and identities in the context of slam competition strategy. The "declaration of self" is also an important concept. A reclamation of stories—especially women's stories—will help to give voice to marginalized experiences in slam poetry. It is my hope through these methods and methodologies that I will be able to shed light on how women (and other marginalized groups) use spoken word poetry as a vehicle for social change and self-healing.
Who, when, and where? : Age-related differences on a novel
Age-related deficits in episodic memory have been documented using well-validated list learning tests such as the California Verbal Learning Test-II (CVLT-II). The present study examined age-related differences on a novel episodic-like memory test assessing memory for "who, when, and where" in addition to associations among these elements. Young (ages 18-25) and older adults (ages 65+) were administered the CVLT-II and a novel episodic-like memory test. Our test consisted of two trials during which the participant was asked to remember a sequence of pictures of different faces paired with different places. The participant then was asked to pair each face with the correct place and put the face-place pairs in the correct sequence. Our test correlated significantly (p < .05) with the CVLT-II, providing preliminary evidence for validity. Older adults remembered significantly fewer face-place pairs and correct pairs in sequence compared to young adults on both trials (ps < .05). Although older adults committed more face and place intrusion errors on the first trial (ps < .05), there were no significant group differences on the second trial. Using a single test, we demonstrated that older adults are impaired in remembering associations between faces and places, as well as the temporal sequence in which face-place pairs were presented; both are critical for everyday episodic memory. Given that intrusion errors for the individual faces/places did not differ on the second trial, these age-related associative memory differences are not due solely to impaired memory for the individual items in the associations.
Whoremongers, heretics, and the devil's doctrine: clerical marriage in mid-Tudor England, 1540-1555
The debate over clerical marriage was a bitter one in the English Protestant Reformation, revealing different views of human nature. For Protestants, the idea that lifelong celibacy could be required of priests was unnatural and harmful. For Catholics, celibacy and chastity was a vow that men could make freely of their own will. This debate over human nature and free will would dominate the clerical marriage controversy during the Reformation in the mid-Tudor period, from the 1530s to the 1550s. The study of clerical marriage is too-often treated in a vacuum, as a theological issue. In fact, clerical marriage debate was, at its basis, as much biological as it was political or ideological. These English theologians attended Cambridge together and knew each other well, but wound up on opposite sides of the debate. This thesis will analyze works by Protestants such as George Joye, John Ponet, and John Hooper, as well as a book written by the Catholic Stephen Gardiner. Joye and Ponet argue that it was human nature to desire to get married and have children. Hooper extends the argument of human nature to encompass God's will, claiming that God would not have created human nature to contradict God's law. In contrast, Gardiner argues that humans had free will, directly contradicting the Protestants' claims. When discussed in their entirety, these works reveal an entirely different analysis of clerical marriage, one that ties into the much larger debate of human nature and free will.
Whose math is it? Building student ownership, agency, and self-efficacy in mathematics
This text serves to assist teachers in promoting student ownership of mathematics by synthesizing relevant research into practical classroom strategies across three tiers of student interaction and experience: classwide, peer-to-peer, and individual. Research from across these tiers of student experience is reviewed and synthesized including literature on developing social and sociomathematical norms, structuring collaborative learning experiences, and promoting self-regulated learning. The concept of success in mathematics is also explored through a review of various policy and consensus documents from the mathematics education community. At its core, individual mathematical ownership is about broadening student-independence and sharpening self- advocacy. Students can be taught to become less reliant on their teachers and ultimately categorize them as just another learning tool. When this happens, when teachers become less needed, mission accomplished. When needs and wants become the limiting factors of one’s mathematical journey, teachers have succeeded – for at that point students truly possess ownership of mathematics. Keywords: Student ownership of mathematics, social norms, sociomathematical norms, teacher clarity, self-regulation of learning, social learning intentions, collaborative learning, study skills, San Diego State University
Why and how the 2018 caravanas from Central America happened and how an Iris Recognition Program could help these massive type of movements to be more efficient and secure for all parties involved
With the increasing influx of migration, much could be done with migration services to assist people who are planning to migrate, including assisting them in gathering the official needed evidence required to qualify for asylum. Similar tools like iris scanning could help uniquely identify people while in their countries of origin and building the unique electronic case for asylum prior to making the extraordinary trip to the US. These same tools can also assist families in tracking their loved ones and interacting with them while in-route to the US or waiting for an asylum hearing in the US. Biometric indicators like an iris scan that uniquely identify an individual and that can be easily obtained in the original countries, as well as being the security and privacy key to developing the asylum package which could make the asylum process far more efficient and secure. Based on the review of literature work on the causes of exodus of people from countries such as Honduras and El Salvador, the authenticity of the eye iris as a way of individual identification, and the current applications of the iris biometric recognition technology in the US and other countries, the analysis shows that the increasing poverty and extreme violence generated by gang members and criminals (belonging to the Mara Salvatrucha or the18 Street gang) are generally credited with being the two main causes for people abandoning these countries, that the iris is the most unique way of identifying an individual, and that in the US iris recognition is currently implemented by many governmental and non-governmental facilities, demonstrating its acceptance. Therefore, using such biometric tools could help avoid multiple risks throughout the entire asylum process, and greatly assisting US Immigration officials by electronically constructing the asylum packages for each person applying for asylum protection. On this basis, a program proposal such as the iRespond is recommended to be implemented in the US. Further research and collaboration are needed to identify other factors that could strengthen the effectiveness of the same., San Diego State University
Why can't people just chill? Factors that encourage and discourage psychological detachment
The percentage of employees working long hours is continuing to increase, leaving little time for employees to effectively recover from the stressors and demands of their work lives. Thousands of studies have documented the negative impacts excessive work stressors or demands can have on both employees' work and nonwork lives. One way for employees to recover from these work stressors or demands is to engage in psychological detachment. Psychological detachment implies not thinking about work, not physically being at work, and not engaging in work-related activities. This study examines the potential social situational influences on an individual employee's detachment. The final sample consisted of responses from 88 full-time employees across a variety of organizations and industries in the United States. Participants self-reported their psychological detachment and provided information about their team-level family supportive supervisor behaviors, masculinity contest culture, and segmentation norms. After checking aggregation statistics, aggregation of family supportive supervisor behaviors, masculinity contest culture, and segmentation norms was not appropriate. At the individual level, only segmentation norms significantly predicted psychological detachment. Future research should obtain a bigger sample size to get enough power to conduct multilevel research. The results of this study suggest that there is some evidence there are social situational influences on individual psychological detachment., San Diego State University
Why so busy? Development of a measure assessing interpretations of coworkers' busyness expressions
In the workplace, it has become increasingly common for people to talk about how busy they are. These busyness expressions and the perceptions surrounding them have received some empirical attention but have not been studied in depth. My thesis takes an inductive, exploratory approach, laying groundwork for what we can learn by studying busyness expressions. I developed the Busyness Expression Interpretation Scale (BEIS) to understand how employees interpret the busyness expressions of those around them. Recruitment via social networking sites, such as Facebook and Linkedin yielded 293 responses that were retained for data analysis. Participants completed a survey distributed via Qualtrics. The survey included the BEIS, as well as measures of public self-consciousness, impression management, work values, job insecurity, subjective job stress, burnout, employee engagement, and job satisfaction. A series of exploratory factor analyses assessed the factor structure of the BEIS, resulting in the removal of five items and partially supporting a three-factor structure of the BEIS, assessing achievement, social, and security interpretation tendencies. Confirmatory factor analysis established the discriminant validity of the BEIS from conceptually related constructs (public self-consciousness, impression management, and work values). Correlations demonstrated significant positive relationships between the BEIS and subjective job stress and burnout, as well as some significant negative relationships between the BEIS and job satisfaction and employee engagement. Participants’ reported frequency of busyness expressions from coworkers and supervisors also demonstrated significant relationships with stress and burnout. Hierarchical regressions demonstrated that the BEIS explains incremental variance over simple frequency in predicting job stress and burnout. The results of these analyses shed light on how busyness expressions in the workplace relate to employees’ well-being. Those who are exposed more to others using busyness expressions may experience more severe stress burnout. How employees interpret busyness expressions provide further insight into how busyness expressions can influence well-being in the form of stress, job satisfaction, engagement, and burnout. Furthermore, the BEIS itself offers an avenue for understanding how busyness expressions operate. Its three- factor structure shows promise but needs to be developed further., San Diego State University
Why we work : a cross-national examination of changing work value orientations and nation-level predictors
A handful of researchers have reported a decline in the work ethic in America following examinations of the nation's workers. This led them to suggest that work values in other countries might be changing, and that patterns in these changes may underlie changes in global competitiveness between nations. However, conclusions supporting a decline in the work ethic in the U.S. are weakened by the fact that they are often derived from responses to the "lottery question." This single-item measure asks subjects how they would behave in a hypothetical situation, and has been shown to be inadequate for assessing a multidimensional construct such as work ethic. Therefore, researchers have called for future studies that assess work value changes over time in a more robust manner. The present study analyzed data collected by the International Social Survey Program (ISSP) across 25 countries between 1989 and 2005, including responses from 89,986 individuals of all ages. Multi-item scales for work centrality and intrinsic/extrinsic work value were constructed, and compared to single-item measures for each construct. Changes over time in these work values were investigated within the U.S. and abroad. Findings suggested that the lottery question may relate to some dimension of work centrality, but generally supported the idea that this item is insufficient for assessing work ethic. Results provided support for a decline in work centrality in the U.S., but suggested a leveling-off of the decline in the lottery question. Finally, the present study explored using changes over time in economic factors (e.g., GDP) to predict changes over time in work values; however, the results yielded limited support for these hypotheses. Overall, the current research provides new insights into the relationships between the lottery question and other important work-related constructs. Findings suggest that there may indeed be a decline in work centrality occurring in the U.S. and in other countries. However, specific reasons for this decline remain unclear and may vary across individuals and systematically across country borders. Future research should further explore this decline, and utilize other data sets to see if these results replicate in other contexts and using other measures.

Pages

Bookmark

Bookmarks: