Matching Items (348)
Filtering by
- All Subjects: artificial intelligence
- Genre: Academic theses
- Creators: Arizona State University
- Resource Type: Text
Description
In this dissertation Mexican American (MA) youths environmental risk contexts, HPA axis functioning and mental health symptomatology were investigated in two separate studies. In the first study, environmental risk contexts were examined utilizing a person-centered approach and focusing on MA adolescents' family, peer, and cultural risk factors in fifth grade (N = 750). Environmental contexts were then linked to mental health symptomatology in seventh grade. Results revealed three distinct environmental contexts: Low risk, Moderate risk-language, and High risk-peer. Youth in the High-risk peer context reported the highest levels of symptomatology; greater major depressive disorder (MDD), anxiety, conduct disorder (CD)/oppositional defiant disorder (ODD), and attention deficit hyperactive disorder (ADHD) symptoms than youth experiencing Low risk or Moderate risk-language context. Females, in particular, experiencing the High risk peer context appeared at greatest risk for MDD symptoms. Finally, adolescents in the Moderate risk-language context displayed similar levels of symptoms to the individuals in the Low risk context, with the exception of higher anxiety. This study suggested that MA youth live in unique environmental contexts and these contexts are differentially related to mental health symptomatology. In the second study, 98 MA youth participated in a three-day diurnal cortisol protocol in hopes of linking perceptions of discrimination and HPA diurnal cortisol rhythms. Results revealed that discrimination was related to greater overall cortisol output and marginally related to the cortisol awakening response and evening levels of cortisol. Results suggest that important physiological processes underlie the experiences of discrimination.
ContributorsZeiders, Katharine H (Author) / Roosa, Mark W. (Thesis advisor) / Doane, Leah D. (Committee member) / Dumka, Larry (Committee member) / Enders, Craig E. (Committee member) / Updegraff, Kimberly A. (Committee member) / Arizona State University (Publisher)
Created2011
Description
Other studies have previously demonstrated that perceived stress and maladaptive stress management can lead to harmful outcomes including depression, morbidity, and mortality. College students (especially freshmen) have more difficulty dealing with stress, which can increase their susceptibility to engage in high risk behaviors. The importance of conducting this research is to discover the effects that perceived stress levels may have on depression outcomes in college students, and to evaluate the influence of health related behaviors on this relationship. This study used a retrospective cross-sectional correlational design to examine correlations between perceived stress, physical activity, and other health behaviors on clinical and perceived depression in college students. A random sample of 20,000 students was drawn from 62,476 students enrolled at Arizona State University (ASU). Participants included 2,238 students who volunteered to take the American College Health Association-National College Health Assessment (ACHA-NCHA) in spring 2009. Supplemental questions for ASU students were developed by ASU Wellness and administered as a part of the ACHA-NCHA II. The university sent an invitation email, wherein students were directed through a hyperlink to the survey website. ACHA provided institutional survey data in an SPSS file for analysis. The data were evaluated with Spearman Rho Correlation Analysis and Wilcoxon-Mann-Whitney test. There were more female participants (n = 580) than males (n = 483), both averaged 23 years of age. Men had greater height, weight, and body mass index than females, all were significant mean differences. There were more significant correlations between health factors and having perceived depression than with having real or diagnosed depression. Logistic regression showed that out of all variables and behaviors studied, only high levels of stress, poor general health, substance use, and gender (female) resulted in significant odds in predicting that a participant would be in one of the depression categories. This research suggests that addressing these factors may be important to prevent and reduce depression among college students. This study provides empirical evidence that there is a significant relationship between perceived stress and depression among college students, and that health behaviors such as substance abuse have a negative mediating effect on this relationship.
ContributorsSkipworth, Katherine (Author) / Swan, Pamela (Thesis advisor) / Woodruff, Larry (Committee member) / Moses, Karen (Committee member) / Arizona State University (Publisher)
Created2011
Description
This dissertation transforms a set of system complexity reduction problems to feature selection problems. Three systems are considered: classification based on association rules, network structure learning, and time series classification. Furthermore, two variable importance measures are proposed to reduce the feature selection bias in tree models. Associative classifiers can achieve high accuracy, but the combination of many rules is difficult to interpret. Rule condition subset selection (RCSS) methods for associative classification are considered. RCSS aims to prune the rule conditions into a subset via feature selection. The subset then can be summarized into rule-based classifiers. Experiments show that classifiers after RCSS can substantially improve the classification interpretability without loss of accuracy. An ensemble feature selection method is proposed to learn Markov blankets for either discrete or continuous networks (without linear, Gaussian assumptions). The method is compared to a Bayesian local structure learning algorithm and to alternative feature selection methods in the causal structure learning problem. Feature selection is also used to enhance the interpretability of time series classification. Existing time series classification algorithms (such as nearest-neighbor with dynamic time warping measures) are accurate but difficult to interpret. This research leverages the time-ordering of the data to extract features, and generates an effective and efficient classifier referred to as a time series forest (TSF). The computational complexity of TSF is only linear in the length of time series, and interpretable features can be extracted. These features can be further reduced, and summarized for even better interpretability. Lastly, two variable importance measures are proposed to reduce the feature selection bias in tree-based ensemble models. It is well known that bias can occur when predictor attributes have different numbers of values. Two methods are proposed to solve the bias problem. One uses an out-of-bag sampling method called OOBForest, and the other, based on the new concept of a partial permutation test, is called a pForest. Experimental results show the existing methods are not always reliable for multi-valued predictors, while the proposed methods have advantages.
ContributorsDeng, Houtao (Author) / Runger, George C. (Thesis advisor) / Lohr, Sharon L (Committee member) / Pan, Rong (Committee member) / Zhang, Muhong (Committee member) / Arizona State University (Publisher)
Created2011
Description
Real-world environments are characterized by non-stationary and continuously evolving data. Learning a classification model on this data would require a framework that is able to adapt itself to newer circumstances. Under such circumstances, transfer learning has come to be a dependable methodology for improving classification performance with reduced training costs and without the need for explicit relearning from scratch. In this thesis, a novel instance transfer technique that adapts a "Cost-sensitive" variation of AdaBoost is presented. The method capitalizes on the theoretical and functional properties of AdaBoost to selectively reuse outdated training instances obtained from a "source" domain to effectively classify unseen instances occurring in a different, but related "target" domain. The algorithm is evaluated on real-world classification problems namely accelerometer based 3D gesture recognition, smart home activity recognition and text categorization. The performance on these datasets is analyzed and evaluated against popular boosting-based instance transfer techniques. In addition, supporting empirical studies, that investigate some of the less explored bottlenecks of boosting based instance transfer methods, are presented, to understand the suitability and effectiveness of this form of knowledge transfer.
ContributorsVenkatesan, Ashok (Author) / Panchanathan, Sethuraman (Thesis advisor) / Li, Baoxin (Committee member) / Ye, Jieping (Committee member) / Arizona State University (Publisher)
Created2011
Description
TaxiWorld is a Matlab simulation of a city with a fleet of taxis which operate within it, with the goal of transporting passengers to their destinations. The size of the city, as well as the number of available taxis and the frequency and general locations of fare appearances can all be set on a scenario-by-scenario basis. The taxis must attempt to service the fares as quickly as possible, by picking each one up and carrying it to its drop-off location. The TaxiWorld scenario is formally modeled using both Decentralized Partially-Observable Markov Decision Processes (Dec-POMDPs) and Multi-agent Markov Decision Processes (MMDPs). The purpose of developing formal models is to learn how to build and use formal Markov models, such as can be given to planners to solve for optimal policies in problem domains. However, finding optimal solutions for Dec-POMDPs is NEXP-Complete, so an empirical algorithm was also developed as an improvement to the method already in use on the simulator, and the methods were compared in identical scenarios to determine which is more effective. The empirical method is of course not optimal - rather, it attempts to simply account for some of the most important factors to achieve an acceptable level of effectiveness while still retaining a reasonable level of computational complexity for online solving.
ContributorsWhite, Christopher (Author) / Kambhampati, Subbarao (Thesis advisor) / Gupta, Sandeep (Committee member) / Varsamopoulos, Georgios (Committee member) / Arizona State University (Publisher)
Created2011
Description
Action language C+ is a formalism for describing properties of actions, which is based on nonmonotonic causal logic. The definite fragment of C+ is implemented in the Causal Calculator (CCalc), which is based on the reduction of nonmonotonic causal logic to propositional logic. This thesis describes the language of CCalc in terms of answer set programming (ASP), based on the translation of nonmonotonic causal logic to formulas under the stable model semantics. I designed a standard library which describes the constructs of the input language of CCalc in terms of ASP, allowing a simple modular method to represent CCalc input programs in the language of ASP. Using the combination of system F2LP and answer set solvers, this method achieves functionality close to that of CCalc while taking advantage of answer set solvers to yield efficient computation that is orders of magnitude faster than CCalc for many benchmark examples. In support of this, I created an automated translation system Cplus2ASP that implements the translation and encoding method and automatically invokes the necessary software to solve the translated input programs.
ContributorsCasolary, Michael (Author) / Lee, Joohyung (Thesis advisor) / Ahn, Gail-Joon (Committee member) / Baral, Chitta (Committee member) / Arizona State University (Publisher)
Created2011
Description
This dissertation explores the lives of women who are on the Severely Mentally Ill (SMI) caseload at Maricopa County Adult Probation in Arizona (The Phoenix metro region). The project focuses on three primary issues: (1) what are the pathways to the criminal justice and mental health systems for women on the SMI caseload (2) how does discretion and expansive formal social control (both benevolent and coercive) impact the lives of these women on the SMI caseload and (3) what are the gendered aspects to successful completion of SMI probation. To answer these questions a mixed-methods research design was employed. First, in-depth semi-structured interviews were completed with 65 women on the SMI caseload. Second, these interviews were supplemented with a case file review of each participant, and field observations (encompassing roughly 100 hours) were conducted at the Maricopa County Mental Health Court. Third, analysis also included 5.5 years of quantitative intake data from the SMI caseload, exploring demographic information and risk and assessment needs scores. The biographies of the women on the SMI caseload revealed similar histories of victimization, substance abuse, and relationship difficulty that previous pathways research has noted. Additionally, mental health problems directly impacted the path to the criminal justice system for some women on the SMI caseload. Results also showed many aspects of expanded social control for women on the SMI caseload. This expanded control appeared to be gendered at times and often created double binds for women. Finally, quantitative analysis showed that some predictive factors of SMI probation completion were gendered. Policy implications and summaries of findings are discussed.
ContributorsMulvey, Philip (Author) / Decker, Scott H. (Thesis advisor) / Spohn, Cassia (Committee member) / Holtfreter, Kristy (Committee member) / Arizona State University (Publisher)
Created2013
Description
Linear Temporal Logic is gaining increasing popularity as a high level specification language for robot motion planning due to its expressive power and scalability of LTL control synthesis algorithms. This formalism, however, requires expert knowledge and makes it inaccessible to non-expert users. This thesis introduces a graphical specification environment to create high level motion plans to control robots in the field by converting a visual representation of the motion/task plan into a Linear Temporal Logic (LTL) specification. The visual interface is built on the Android tablet platform and provides functionality to create task plans through a set of well defined gestures and on screen controls. It uses the notion of waypoints to quickly and efficiently describe the motion plan and enables a variety of complex Linear Temporal Logic specifications to be described succinctly and intuitively by the user without the need for the knowledge and understanding of LTL specification. Thus, it opens avenues for its use by personnel in military, warehouse management, and search and rescue missions. This thesis describes the construction of LTL for various scenarios used for robot navigation using the visual interface developed and leverages the use of existing LTL based motion planners to carry out the task plan by a robot.
ContributorsSrinivas, Shashank (Author) / Fainekos, Georgios (Thesis advisor) / Baral, Chitta (Committee member) / Burleson, Winslow (Committee member) / Arizona State University (Publisher)
Created2013
High school students' perceptions of teaching and their intention to choose teaching as a profession
Description
This study was conducted to (a) explore high achieving high school students' perceptions of the teaching profession, (b) examine the influence of these perceptions on intentions to teach, and (c) test a recruitment suite of tools to determine the effectiveness of recruitment messaging and strategies. The Theory of Planned Behavior (TPB) served as the theoretical framework for this study. Using the TPB allowed examination of students' behavioral, normative, and control beliefs as well as their attitudes, subjective norms, and efficacy and how those components affected intentions to teach. Participants included high school seniors in the top 20% of their class. A mixed methods approach was employed to identify how the characteristics that students value when considering a profession were aligned with those they believed to be true about the teaching profession. Additionally mixing methods allowed for a more thorough exploration of the matter and an in-depth depiction of perceptions and intentions to teach. Results from a confirmatory path analysis showed students' perceived behavioral control, a measure of efficacy, and attitudes toward teaching were predictive of intention to teach and accounted for 25% of the variation in intention to teach scores. A series of exploratory structural equation models was developed to examine additional paths that might be useful in understanding students' intention to teach. Three additional, important paths were found among TPB variables that accounted for an additional 14% of the variation in intention scores. Additionally, these paths had implications for recruitment practice. Five themes emerged from the qualitative data--status, societal importance, influences of important others, teaching as a backup option, and barriers. The discussion focused on implications for recruitment practice and research, limitations, and conclusions. The following conclusions were drawn: (a) students must be provided with knowledge about the teaching profession to overcome stereotypical beliefs, (b) recruitment must begin much earlier, (c) parents must be better informed about teaching, (d) use of a longer recruitment process with multiple touch points must be used to inform and inspire students, and (e) students must be provided with practice teaching opportunities and systematic observational opportunities, which can foster increased efficacy for teaching.
ContributorsCruz, Crystal (Author) / Buss, Ray R (Thesis advisor) / Barnett, Joshua (Committee member) / Bentz, Matthew (Committee member) / Arizona State University (Publisher)
Created2013
Description
Answer Set Programming (ASP) is one of the most prominent and successful knowledge representation paradigms. The success of ASP is due to its expressive non-monotonic modeling language and its efficient computational methods originating from building propositional satisfiability solvers. The wide adoption of ASP has motivated several extensions to its modeling language in order to enhance expressivity, such as incorporating aggregates and interfaces with ontologies. Also, in order to overcome the grounding bottleneck of computation in ASP, there are increasing interests in integrating ASP with other computing paradigms, such as Constraint Programming (CP) and Satisfiability Modulo Theories (SMT). Due to the non-monotonic nature of the ASP semantics, such enhancements turned out to be non-trivial and the existing extensions are not fully satisfactory. We observe that one main reason for the difficulties rooted in the propositional semantics of ASP, which is limited in handling first-order constructs (such as aggregates and ontologies) and functions (such as constraint variables in CP and SMT) in natural ways. This dissertation presents a unifying view on these extensions by viewing them as instances of formulas with generalized quantifiers and intensional functions. We extend the first-order stable model semantics by by Ferraris, Lee, and Lifschitz to allow generalized quantifiers, which cover aggregate, DL-atoms, constraints and SMT theory atoms as special cases. Using this unifying framework, we study and relate different extensions of ASP. We also present a tight integration of ASP with SMT, based on which we enhance action language C+ to handle reasoning about continuous changes. Our framework yields a systematic approach to study and extend non-monotonic languages.
ContributorsMeng, Yunsong (Author) / Lee, Joohyung (Thesis advisor) / Ahn, Gail-Joon (Committee member) / Baral, Chitta (Committee member) / Fainekos, Georgios (Committee member) / Lifschitz, Vladimir (Committee member) / Arizona State University (Publisher)
Created2013