Matching Items (703)
Filtering by

Clear all filters

158399-Thumbnail Image.png
Description
Languages, specially gestural and sign languages, are best learned in immersive environments with rich feedback. Computer-Aided Language Learning (CALL) solu- tions for spoken languages have successfully incorporated some feedback mechanisms, but no such solution exists for signed languages. Computer Aided Sign Language Learning (CASLL) is a recent and promising field

Languages, specially gestural and sign languages, are best learned in immersive environments with rich feedback. Computer-Aided Language Learning (CALL) solu- tions for spoken languages have successfully incorporated some feedback mechanisms, but no such solution exists for signed languages. Computer Aided Sign Language Learning (CASLL) is a recent and promising field of research which is made feasible by advances in Computer Vision and Sign Language Recognition(SLR). Leveraging existing SLR systems for feedback based learning is not feasible because their decision processes are not human interpretable and do not facilitate conceptual feedback to learners. Thus, fundamental research is needed towards designing systems that are modular and explainable. The explanations from these systems can then be used to produce feedback to aid in the learning process.

In this work, I present novel approaches for the recognition of location, movement and handshape that are components of American Sign Language (ASL) using both wrist-worn sensors as well as webcams. Finally, I present Learn2Sign(L2S), a chat- bot based AI tutor that can provide fine-grained conceptual feedback to learners of ASL using the modular recognition approaches. L2S is designed to provide feedback directly relating to the fundamental concepts of ASL using an explainable AI. I present the system performance results in terms of Precision, Recall and F-1 scores as well as validation results towards the learning outcomes of users. Both retention and execution tests for 26 participants for 14 different ASL words learned using learn2sign is presented. Finally, I also present the results of a post-usage usability survey for all the participants. In this work, I found that learners who received live feedback on their executions improved their execution as well as retention performances. The average increase in execution performance was 28% points and that for retention was 4% points.
ContributorsPaudyal, Prajwal (Author) / Gupta, Sandeep (Thesis advisor) / Banerjee, Ayan (Committee member) / Hsiao, Ihan (Committee member) / Azuma, Tamiko (Committee member) / Yang, Yezhou (Committee member) / Arizona State University (Publisher)
Created2020
158566-Thumbnail Image.png
Description
Social media has become an important means of user-centered information sharing and communications in a gamut of domains, including news consumption, entertainment, marketing, public relations, and many more. The low cost, easy access, and rapid dissemination of information on social media draws a large audience but also exacerbate the wide

Social media has become an important means of user-centered information sharing and communications in a gamut of domains, including news consumption, entertainment, marketing, public relations, and many more. The low cost, easy access, and rapid dissemination of information on social media draws a large audience but also exacerbate the wide propagation of disinformation including fake news, i.e., news with intentionally false information. Disinformation on social media is growing fast in volume and can have detrimental societal effects. Despite the importance of this problem, our understanding of disinformation in social media is still limited. Recent advancements of computational approaches on detecting disinformation and fake news have shown some early promising results. Novel challenges are still abundant due to its complexity, diversity, dynamics, multi-modality, and costs of fact-checking or annotation.

Social media data opens the door to interdisciplinary research and allows one to collectively study large-scale human behaviors otherwise impossible. For example, user engagements over information such as news articles, including posting about, commenting on, or recommending the news on social media, contain abundant rich information. Since social media data is big, incomplete, noisy, unstructured, with abundant social relations, solely relying on user engagements can be sensitive to noisy user feedback. To alleviate the problem of limited labeled data, it is important to combine contents and this new (but weak) type of information as supervision signals, i.e., weak social supervision, to advance fake news detection.

The goal of this dissertation is to understand disinformation by proposing and exploiting weak social supervision for learning with little labeled data and effectively detect disinformation via innovative research and novel computational methods. In particular, I investigate learning with weak social supervision for understanding disinformation with the following computational tasks: bringing the heterogeneous social context as auxiliary information for effective fake news detection; discovering explanations of fake news from social media for explainable fake news detection; modeling multi-source of weak social supervision for early fake news detection; and transferring knowledge across domains with adversarial machine learning for cross-domain fake news detection. The findings of the dissertation significantly expand the boundaries of disinformation research and establish a novel paradigm of learning with weak social supervision that has important implications in broad applications in social media.
ContributorsShu, Kai (Author) / Liu, Huan (Thesis advisor) / Bernard, H. Russell (Committee member) / Maciejewski, Ross (Committee member) / Xue, Guoliang (Committee member) / Arizona State University (Publisher)
Created2020
158534-Thumbnail Image.png
Description
ABSTRACT

Background. College students’ modifiable health behaviors, including unhealthful eating patterns, predispose them to risk for future cardiometabolic conditions.

Purpose. This novel 8-week randomized control parallel-arm study compared the effects of a daily 18-hour Time-Restricted Feeding protocol vs. an 8-hour fast on diet quality in college students. Secondary outcomes were resting

ABSTRACT

Background. College students’ modifiable health behaviors, including unhealthful eating patterns, predispose them to risk for future cardiometabolic conditions.

Purpose. This novel 8-week randomized control parallel-arm study compared the effects of a daily 18-hour Time-Restricted Feeding protocol vs. an 8-hour fast on diet quality in college students. Secondary outcomes were resting morning blood pressure, biomarkers of glucose regulation, biomarkers of lipid metabolism, and anthropometric measures.

Methods. Eighteen healthy college students (age = 23 ± 4 years; BMI = 23.2 ± 2.3 kg/m2; MET = 58.8 ± 32.9 min/wk) completed this study. Participants were randomized to a daily 18-hour fasting protocol (Intervention; n = 8) or a daily 8-hour fasting protocol (Control; n = 10) for eight weeks. One ‘cheat’ day was permitted each week. Outcomes were measured at weeks 0 (baseline), 4, and 8. A non-parametric Mann Whitney U test was used to compare the week 4 change from baseline between groups. Statistical significance was set at p≤0.05.

Results. Diet quality (p = 0.030) and body weight (p = 0.016) improved from baseline to week 4 for the INV group in comparison to the CON group. The data suggest these improvements may be related to reductions in snacking frequency and increased breakfast consumption. Fasting blood glucose and hip circumference tended to improve for the INV group in comparison to the CON group (p = 0.091 and p = 0.100). However, saturated fat intake tended to increase in the INV group in comparison to the CON group (p = 0.064). Finally, there were no treatment differences between groups (p>0.05) for the 4-week change in total calories, dietary vitamin C, added sugars, resting systolic blood pressure, resting diastolic blood pressure, insulin, homeostatic model assessment for insulin resistance (HOMA-IR), low-density lipoprotein (LDL) cholesterol, triglycerides, high-density lipoprotein (HDL) cholesterol, waist circumference, or MET.

Conclusion. These data, although preliminary, suggest that the 18-hour fasting protocol was effective for improving diet quality and reducing weight in comparison to the 8-hour fasting protocol in healthy college students. Future intervention trials will need to confirm these findings and determine the long-term relevance of these improvements for health outcomes.
ContributorsMayra, Selicia (Author) / Johnston, Carol (Thesis advisor) / Sears, Dorothy (Committee member) / Swan, Pamela (Committee member) / Sweazea, Karen (Committee member) / Wharton, Christopher (Christopher Mack), 1977- (Committee member) / Arizona State University (Publisher)
Created2020
158552-Thumbnail Image.png
Description
The recent increase in users of cellular networks necessitates the use of new technologies to meet this demand. Massive multiple input multiple output (MIMO) communication systems have great potential for increasing the network capacity of the emerging 5G+ cellular networks. However, leveraging the multiplexing and beamforming gains from these large-scale

The recent increase in users of cellular networks necessitates the use of new technologies to meet this demand. Massive multiple input multiple output (MIMO) communication systems have great potential for increasing the network capacity of the emerging 5G+ cellular networks. However, leveraging the multiplexing and beamforming gains from these large-scale MIMO systems requires the channel knowlege between each antenna and each user. Obtaining channel information on such a massive scale is not feasible with the current technology available due to the complexity of such large systems. Recent research shows that deep learning methods can lead to interesting gains for massive MIMO systems by mapping the channel information from the uplink frequency band to the channel information for the downlink frequency band as well as between antennas at nearby locations. This thesis presents the research to develop a deep learning based channel mapping proof-of-concept prototype.



Due to deep neural networks' need of large training sets for accurate performance, this thesis outlines the design and implementation of an autonomous channel measurement system to analyze the performance of the proposed deep learning based channel mapping concept. This system obtains channel magnitude measurements from eight antennas autonomously using a mobile robot carrying a transmitter which receives wireless commands from the central computer connected to the static receiver system. The developed autonomous channel measurement system is capable of obtaining accurate and repeatable channel magnitude measurements. It is shown that the proposed deep learning based channel mapping system accurately predicts channel information containing few multi-path effects.
ContributorsBooth, Jayden Charles (Author) / Spanias, Andreas (Thesis advisor) / Alkhateeb, Ahmed (Thesis advisor) / Ewaisha, Ahmed (Committee member) / Arizona State University (Publisher)
Created2020
Description
Movement disorders are becoming one of the leading causes of functional disability due to aging populations and extended life expectancy. Diagnosis, treatment, and rehabilitation currently depend on the behavior observed in a clinical environment. After the patient leaves the clinic, there is no standard approach to continuously monitor the patient

Movement disorders are becoming one of the leading causes of functional disability due to aging populations and extended life expectancy. Diagnosis, treatment, and rehabilitation currently depend on the behavior observed in a clinical environment. After the patient leaves the clinic, there is no standard approach to continuously monitor the patient and report potential problems. Furthermore, self-recording is inconvenient and unreliable. To address these challenges, wearable health monitoring is emerging as an effective way to augment clinical care for movement disorders.

Wearable devices are being used in many health, fitness, and activity monitoring applications. However, their widespread adoption has been hindered by several adaptation and technical challenges. First, conventional rigid devices are uncomfortable to wear for long periods. Second, wearable devices must operate under very low-energy budgets due to their small battery capacities. Small batteries create a need for frequent recharging, which in turn leads users to stop using them. Third, the usefulness of wearable devices must be demonstrated through high impact applications such that users can get value out of them.

This dissertation presents solutions to solving the challenges faced by wearable devices. First, it presents an open-source hardware/software platform for wearable health monitoring. The proposed platform uses flexible hybrid electronics to enable devices that conform to the shape of the user’s body. Second, it proposes an algorithm to enable recharge-free operation of wearable devices that harvest energy from the environment. The proposed solution maximizes the performance of the wearable device under minimum energy constraints. The results of the proposed algorithm are, on average, within 3% of the optimal solution computed offline. Third, a comprehensive framework for human activity recognition (HAR), one of the first steps towards a solution for movement disorders is presented. It starts with an online learning framework for HAR. Experiments on a low power IoT device (TI-CC2650 MCU) with twenty-two users show 95% accuracy in identifying seven activities and their transitions with less than 12.5 mW power consumption. The online learning framework is accompanied by a transfer learning approach for HAR that determines the number of neural network layers to transfer among uses to enable efficient online learning. Next, a technique to co-optimize the accuracy and active time of wearable applications by utilizing multiple design points with different energy-accuracy trade-offs is presented. The proposed technique switches between the design points at runtime to maximize a generalized objective function under tight harvested energy budget constraints. Finally, we present the first ultra-low-energy hardware accelerator that makes it practical to perform HAR on energy harvested from wearable devices. The accelerator consumes 22.4 microjoules per operation using a commercial 65 nm technology. In summary, the solutions presented in this dissertation can enable the wider adoption of wearable devices.
ContributorsBhat, Ganapati (Author) / Ogras, Umit Y. (Thesis advisor) / Chakrabarti, Chaitali (Committee member) / Nedić, Angelia (Committee member) / Marculescu, Radu (Committee member) / Arizona State University (Publisher)
Created2020
158561-Thumbnail Image.png
Description
The assessment and evaluation of dietary intake and nutrition knowledge in female athletes is especially important due to the high prevalence of inadequate intake in this population (Black et al., 2019). This study evaluated 1) the relationship of nutrition knowledge and dietary intake among collegiate female athletes at the National

The assessment and evaluation of dietary intake and nutrition knowledge in female athletes is especially important due to the high prevalence of inadequate intake in this population (Black et al., 2019). This study evaluated 1) the relationship of nutrition knowledge and dietary intake among collegiate female athletes at the National Collegiate Athletic Association (NCAA) Division I, National Junior College Athletic Association (NJCAA), and Club sport levels and 2) the impact of competition level on this relationship as well. Participants (NCAA DI, n=51; NJCAA, n = 36; Club, n = 37) in this study answered two questionnaires, the Nutrition Sport Knowledge Questionnaire (NSKQ) and the Rapid Eating Assessment for Participants (REAP) questionnaire to assess knowledge and dietary intake. Participants also provided anthropometric and demographic information. The NSKQ was scored as a whole and for each of the four subcategories. REAP was scored both by tallying the number of “usually/often” frequency responses and given a numeral score to estimate diet quality. Statistical analysis was conducted using Kruskal-Wallis, Chi-square and Spearman’s correlation tests to compare differences within subgroups of participants and evaluate any relationships that may exist between nutrition knowledge and dietary intake with significance set at p≤0.05. Differences in nutrition knowledge between competition groups were significant, H(2)= 16.94, p< 0.001. NCAA DI (p<0.001) and Club (p<0.001) athletes had higher nutrition knowledge than athletes at the NJCAA level. This was true for overall knowledge as well as knowledge subcategories. However, minimal relationships between nutrition knowledge and dietary intake were found. The overall correlation value was rs(118)= -0.10 (95%CI: -0.28 to 0.08), p>0.05. This suggests those with higher nutrition knowledge did not necessarily have better dietary intake. Improvements in the assessment of nutrition knowledge and quick assessment of dietary quality and the relation between both is needed.
ContributorsSkinner, Jensen Drew (Author) / Wardenaar, Floris (Thesis advisor) / Johnston, Carol (Committee member) / Yudell, Amber (Committee member) / Arizona State University (Publisher)
Created2020
158239-Thumbnail Image.png
Description
The Adequate Intake (AI) level for total fiber for adults is 14 grams per 1,000 kilocalories per day; however, only 12.9% of Americans met their total fiber needs according to the 2015-2016 National Health and Nutrition Examination Survey (NHANES). A lower frequency of home-cooked meals and a higher frequency of

The Adequate Intake (AI) level for total fiber for adults is 14 grams per 1,000 kilocalories per day; however, only 12.9% of Americans met their total fiber needs according to the 2015-2016 National Health and Nutrition Examination Survey (NHANES). A lower frequency of home-cooked meals and a higher frequency of restaurant meals have been cited as a possible explanation for the low dietary fiber intake among Americans, and according to the Social-Ecological Model, the retail food environment can influence our food choices such as the choice to eat at home or eat out. The objective of this study is to examine the relationship between a dynamic measurement of exposure to the retail food environment and fiber intake (total fiber, soluble fiber, insoluble fiber, and pectin). This is a secondary analysis of data from the Community of Mine study, a cross-sectional study of 602 adults residing in San Diego County, California. Dynamic exposure to the retail food environment was assessed using Global Positioning Systems (GPS) data collected by the Qstarz GPS device worn by each participant. Fiber intake was assessed using two 24-hour dietary recalls. Multivariate regression analysis was used to assess correlations. Descriptive results showed no significant differences in dynamic exposure to the retail food environment by sex, Hispanic ethnicity, and income. There were significant differences in fiber intake by sex and ethnicity. The results of the multivariate regression analysis suggest that exposure to the retail food environment is not associated with fiber intake among a subset of American adults.
ContributorsHarb, Amanda A (Author) / Sears, Dorothy (Thesis advisor) / Alexon, Christy (Committee member) / Jankowska, Marta (Committee member) / Ohri-Vachaspati, Punam (Committee member) / Arizona State University (Publisher)
Created2020
158249-Thumbnail Image.png
Description
Introduction: A diet high in fermented, oligio-, di-, monosaccharide, and polyols

(FODMAP) has been shown to exacerbate symptoms of irritable bowel syndrome

(IBS). Previous literature has shown significant improvement in IBS symptoms after

adherence to a low FODMAP diet (LFD). However, dietary adherence to the LFD is

difficult with patients stating that information provided

Introduction: A diet high in fermented, oligio-, di-, monosaccharide, and polyols

(FODMAP) has been shown to exacerbate symptoms of irritable bowel syndrome

(IBS). Previous literature has shown significant improvement in IBS symptoms after

adherence to a low FODMAP diet (LFD). However, dietary adherence to the LFD is

difficult with patients stating that information provided by healthcare providers

(HCPs) is generalized and nonspecific requiring them to search for supplementary

information to fit their needs. Notably, studies that have used a combination of

online and in-person methods for treatment have shown improved adherence to the

LFD. Objective: To determine whether a novel artificial intelligence (AI) dietary

mobile application will improve adherence to the LFD compared to a standard online

dietary intervention (CON) in populations with IBS or IBS-like symptoms over a 4-

week period. Methods: Participants were randomized into two groups: APP or CON.

The intervention group was provided access to an AI mobile application, a dietary

resource verified by registered dietitians which uses artificial intelligence to

individualize dietary guidance in real-time with the ability to scan menus and

barcodes and provide individuals with food scores based on their dietary preferences.

Primary measures included mobile app engagement, dietary adherence, and

manifestation of IBS-like symptoms. Baseline Results: A total of 58 participants

were randomized to groups. This is an ongoing study and this thesis details the

methodology and baseline characteristics of the participants at baseline and

intervention start. Validation of the application could improve the range of offerings

for lifestyle diseases treatable through dietary modification.
ContributorsRafferty, Aaron (Author) / Johnston, Carol (Thesis advisor) / Hall, Richard (Committee member) / Fitton, Renee (Committee member) / Arizona State University (Publisher)
Created2020
158259-Thumbnail Image.png
Description
In the last decade deep learning based models have revolutionized machine learning and computer vision applications. However, these models are data-hungry and training them is a time-consuming process. In addition, when deep neural networks are updated to augment their prediction space with new data, they run into the problem of

In the last decade deep learning based models have revolutionized machine learning and computer vision applications. However, these models are data-hungry and training them is a time-consuming process. In addition, when deep neural networks are updated to augment their prediction space with new data, they run into the problem of catastrophic forgetting, where the model forgets previously learned knowledge as it overfits to the newly available data. Incremental learning algorithms enable deep neural networks to prevent catastrophic forgetting by retaining knowledge of previously observed data while also learning from newly available data.

This thesis presents three models for incremental learning; (i) Design of an algorithm for generative incremental learning using a pre-trained deep neural network classifier; (ii) Development of a hashing based clustering algorithm for efficient incremental learning; (iii) Design of a student-teacher coupled neural network to distill knowledge for incremental learning. The proposed algorithms were evaluated using popular vision datasets for classification tasks. The thesis concludes with a discussion about the feasibility of using these techniques to transfer information between networks and also for incremental learning applications.
ContributorsPatil, Rishabh (Author) / Venkateswara, Hemanth (Thesis advisor) / Panchanathan, Sethuraman (Thesis advisor) / McDaniel, Troy (Committee member) / Arizona State University (Publisher)
Created2020
Description
Hyperspectral unmixing is an important remote sensing task with applications including material identification and analysis. Characteristic spectral features make many pure materials identifiable from their visible-to-infrared spectra, but quantifying their presence within a mixture is a challenging task due to nonlinearities and factors of variation. In this thesis, physics-based approaches

Hyperspectral unmixing is an important remote sensing task with applications including material identification and analysis. Characteristic spectral features make many pure materials identifiable from their visible-to-infrared spectra, but quantifying their presence within a mixture is a challenging task due to nonlinearities and factors of variation. In this thesis, physics-based approaches are incorporated into an end-to-end spectral unmixing algorithm via differentiable programming. First, sparse regularization and constraints are implemented by adding differentiable penalty terms to a cost function to avoid unrealistic predictions. Secondly, a physics-based dispersion model is introduced to simulate realistic spectral variation, and an efficient method to fit the parameters is presented. Then, this dispersion model is utilized as a generative model within an analysis-by-synthesis spectral unmixing algorithm. Further, a technique for inverse rendering using a convolutional neural network to predict parameters of the generative model is introduced to enhance performance and speed when training data are available. Results achieve state-of-the-art on both infrared and visible-to-near-infrared (VNIR) datasets as compared to baselines, and show promise for the synergy between physics-based models and deep learning in hyperspectral unmixing in the future.
ContributorsJaniczek, John (Author) / Jayasuriya, Suren (Thesis advisor) / Dasarathy, Gautam (Thesis advisor) / Christensen, Phil (Committee member) / Arizona State University (Publisher)
Created2020