Statistical dialog management for health interventions

Research endeavors on spoken dialogue systems in the 1990s and 2000s have led to the deployment of commercial spoken dialogue systems (SDS) in microdomains such as customer service automation, reservation/booking and question answering systems. Recent research in SDS has been focused on the development of applications in different domains (e.g. virtual counseling, personal coaches, social companions) which requires more sophistication than the previous generation of commercial SDS. The focus of this research project is the delivery of behavior change interventions based on the brief intervention counseling style via spoken dialogue systems. ^ Brief interventions (BI) are evidence-based, short, well structured, one-on-one counseling sessions. Many challenges are involved in delivering BIs to people in need, such as finding the time to administer them in busy doctors' offices, obtaining the extra training that helps staff become comfortable providing these interventions, and managing the cost of delivering the interventions. Fortunately, recent developments in spoken dialogue systems make the development of systems that can deliver brief interventions possible. ^ The overall objective of this research is to develop a data-driven, adaptable dialogue system for brief interventions for problematic drinking behavior, based on reinforcement learning methods. The implications of this research project includes, but are not limited to, assessing the feasibility of delivering structured brief health interventions with a data-driven spoken dialogue system. Furthermore, while the experimental system focuses on harmful alcohol drinking as a target behavior in this project, the produced knowledge and experience may also lead to implementation of similarly structured health interventions and assessments other than the alcohol domain (e.g. obesity, drug use, lack of exercise), using statistical machine learning approaches. ^ In addition to designing a dialog system, the semantic and emotional meanings of user utterances have high impact on interaction. To perform domain specific reasoning and recognize concepts in user utterances, a named-entity recognizer and an ontology are designed and evaluated. To understand affective information conveyed through text, lexicons and sentiment analysis module are developed and tested.^

Division of Research Annual Customer Survey

Customer survey results for 2012-2013.

Labeling large scale social media data using budget-driven One-class SVM classification

The social media classification problems draw more and more attention in the past few years. With the rapid development of Internet and the popularity of computers, there is astronomical amount of information in the social network (social media platforms). The datasets are generally large scale and are often corrupted by noise. The presence of noise in training set has strong impact on the performance of supervised learning (classification) techniques. A budget-driven One-class SVM approach is presented in this thesis that is suitable for large scale social media data classification. Our approach is based on an existing online One-class SVM learning algorithm, referred as STOCS (Self-Tuning One-Class SVM) algorithm. To justify our choice, we first analyze the noise-resilient ability of STOCS using synthetic data. The experiments suggest that STOCS is more robust against label noise than several other existing approaches. Next, to handle big data classification problem for social media data, we introduce several budget driven features, which allow the algorithm to be trained within limited time and under limited memory requirement. Besides, the resulting algorithm can be easily adapted to changes in dynamic data with minimal computational cost. Compared with two state-of-the-art approaches, Lib-Linear and kNN, our approach is shown to be competitive with lower requirements of memory and time.

Development of a high-frequency surface-wave radar cross section model for icebergs

In this thesis, the first-order radar cross section (RCS) of an iceberg is derived and simulated. This analysis takes place in the context of a monostatic high frequency surface wave radar with a vertical dipole source that is driven by a pulsed waveform. The starting point of this work is a general electric field equation derived previ- ously for an arbitrarily shaped iceberg region surrounded by an ocean surface. The condition of monostatic backscatter is applied to this general field equation and the resulting expression is inverse Fourier transformed. In the time domain the excitation current of the transmit antenna is specified to be a pulsed sinusoid signal. The result- ing electric field equation is simplified and its physical significance is assessed. The field equation is then further simplified by restricting the iceberg's size to fit within a single radar patch width. The power received by the radar is calculated using this electric field equation. Comparing the received power with the radar range equation gives a general expression for the iceberg RCS. The iceberg RCS equation is found to depend on several parameters including the geometry of the iceberg, the radar frequency, and the electrical parameters of both the iceberg and the ocean surface. The RCS is rewritten in a form suitable for simulations and simulations are carried out for rectangularly shaped icebergs. Simulation results are discussed and are found to be consistent with existing research.

E-service quality:development of a hierarchical model

Using survey data from 358 online customers, the study finds that the e-service quality construct conforms to the structure of a third-order factor model that links online service quality perceptions to distinct and actionable dimensions, including (1) website design, (2) fulfilment, (3) customer service, and (4) security/privacy. Each dimension is found to consist of several attributes that define the basis of e-service quality perceptions. A comprehensive specification of the construct, which includes attributes not covered in existing scales, is developed. The study contrasts a formative model consisting of 4 dimensions and 16 attributes against a reflective conceptualization. The results of this comparison indicate that studies using an incorrectly specified model overestimate the importance of certain e-service quality attributes. Global fit criteria are also found to support the detection of measurement misspecification. Meta-analytic data from 31,264 online customers are used to show that the developed measurement predicts customer behavior better than widely used scales, such as WebQual and E-S-Qual. The results show that the new measurement enables managers to assess e-service quality more accurately and predict customer behavior more reliably.

Colorado's Piceance Basin: Variation in the local public finance effects of oil and gas development

A large increase in natural gas production occurred in western Colorado’s Piceance basin in the mid- to late-2000s, generating a surge in population, economic activity, and heavy truck traffic in this rural region. We describe the fiscal effects related to this development for two county governments: Garfield and Rio Blanco, and two city governments: Grand Junction and Rifle. Counties maintain rural road networks in Colorado, and Garfield County’s ability to fashion agreements with operators to repair roads damaged during operations helped prevent the types of large new costs seen in Rio Blanco County, a neighboring county with less government capacity and where such agreements were not made. Rifle and Grand Junction experienced substantial oil- and gas-driven population growth, with greater challenges in the smaller, more isolated, and less economically diverse city of Rifle. Lessons from this case study include the value of crafting road maintenance agreements, fiscal risks for small and geographically isolated communities experiencing rapid population growth, challenges associated with limited infrastructure, and the desirability of flexibility in the allocation of oil- and gas-related revenue.

Global Mental Health: Five Areas for Value-Driven Training Innovation.

OBJECTIVE: In the field of global mental health, there is a need for identifying core values and competencies to guide training programs in professional practice as well as in academia. This paper presents the results of interdisciplinary discussions fostered during an annual meeting of the Society for the Study of Psychiatry and Culture to develop recommendations for value-driven innovation in global mental health training. METHODS: Participants (n = 48), who registered for a dedicated workshop on global mental health training advertised in conference proceedings, included both established faculty and current students engaged in learning, practice, and research. They proffered recommendations in five areas of training curriculum: values, competencies, training experiences, resources, and evaluation. RESULTS: Priority values included humility, ethical awareness of power differentials, collaborative action, and "deep accountability" when working in low-resource settings in low- and middle-income countries and high-income countries. Competencies included flexibility and tolerating ambiguity when working across diverse settings, the ability to systematically evaluate personal biases, historical and linguistic proficiency, and evaluation skills across a range of stakeholders. Training experiences included didactics, language training, self-awareness, and supervision in immersive activities related to professional or academic work. Resources included connections with diverse faculty such as social scientists and mentors in addition to medical practitioners, institutional commitment through protected time and funding, and sustainable collaborations with partners in low resource settings. Finally, evaluation skills built upon community-based participatory methods, 360-degree feedback from partners in low-resource settings, and observed structured clinical evaluations (OSCEs) with people of different cultural backgrounds. CONCLUSIONS: Global mental health training, as envisioned in this workshop, exemplifies an ethos of working through power differentials across clinical, professional, and social contexts in order to form longstanding collaborations. If incorporated into the ACGME/ABPN Psychiatry Milestone Project, such recommendations will improve training gained through international experiences as well as the everyday training of mental health professionals, global health practitioners, and social scientists.

Development and Optimization of Four-dimensional Magnetic Resonance Imaging (4D-MRI) for Radiation Therapy

A tenet of modern radiotherapy (RT) is to identify the treatment target accurately, following which the high-dose treatment volume may be expanded into the surrounding tissues in order to create the clinical and planning target volumes. Respiratory motion can induce errors in target volume delineation and dose delivery in radiation therapy for thoracic and abdominal cancers. Historically, radiotherapy treatment planning in the thoracic and abdominal regions has used 2D or 3D images acquired under uncoached free-breathing conditions, irrespective of whether the target tumor is moving or not. Once the gross target volume has been delineated, standard margins are commonly added in order to account for motion. However, the generic margins do not usually take the target motion trajectory into consideration. That may lead to under- or over-estimate motion with subsequent risk of missing the target during treatment or irradiating excessive normal tissue. That introduces systematic errors into treatment planning and delivery. In clinical practice, four-dimensional (4D) imaging has been popular in For RT motion management. It provides temporal information about tumor and organ at risk motion, and it permits patient-specific treatment planning. The most common contemporary imaging technique for identifying tumor motion is 4D computed tomography (4D-CT). However, CT has poor soft tissue contrast and it induce ionizing radiation hazard. In the last decade, 4D magnetic resonance imaging (4D-MRI) has become an emerging tool to image respiratory motion, especially in the abdomen, because of the superior soft-tissue contrast. Recently, several 4D-MRI techniques have been proposed, including prospective and retrospective approaches. Nevertheless, 4D-MRI techniques are faced with several challenges: 1) suboptimal and inconsistent tumor contrast with large inter-patient variation; 2) relatively low temporal-spatial resolution; 3) it lacks a reliable respiratory surrogate. In this research work, novel 4D-MRI techniques applying MRI weightings that was not used in existing 4D-MRI techniques, including T2/T1-weighted, T2-weighted and Diffusion-weighted MRI were investigated. A result-driven phase retrospective sorting method was proposed, and it was applied to image space as well as k-space of MR imaging. Novel image-based respiratory surrogates were developed, improved and evaluated.

Local fiscal effects of oil and gas development in eight states

Oil and gas production in the United States has increased dramatically in the past 10 years. This growth has important implications for local governments, which often see new revenues from a variety of sources: property taxes on oil and gas property, sales taxes driven by the oil and gas workforce, allocations of state revenues from severance taxes or state and federal leases, leases on local government land, and contributions from oil and gas companies to support local services. At the same time, local governments tend to experience a range of new costs such as road damage caused by heavy industry truck traffic, increased demand for emergency services and law enforcement, and challenges with workforce retention. This report examines county and municipal fiscal effects in 14 oil- and gas-producing regions of eight states: AK, CA, KS, OH, OK, NM, UT, and WV. We find that for most local governments, oil and gas development—whether new or longstanding—has a positive effect on local public finances. However, effects can vary substantially due to a variety of local factors and policy issues. For some local governments, particularly those in rural regions experiencing large increases in development, revenues have not kept pace with rapidly increased costs and demand for services, particularly on road repair.

Through the Lenses of Q’eqchi Maya: (Re)Framing the Story of Development in a Guatemalan Indigenous Community Through Participant-Created Photographs

The question of evaluations of development projects has been widely debated within the field of international development, with scholars and development practitioners calling for increased community-driven evaluations. However, there has been a paucity of research in community-led project evaluations, and a largely absent investigation utilizing visual anthropology/sociology methodologies. This paper seeks to shift this power by giving voice to the intended beneficiaries of an eco-tourism project in a rural indigenous Guatemala village. Through photographs taken by community members and corresponding interviews, this paper shows the way in which community members have and continue to reframe the idea of development in their village. Specifically, my analysis reveals how residents see changing forms of access, how they reframe ideas of beauty and modernization, and how they reframe their relationship to the land through Western conservation and private property ideals. This research thus provides an alternative narrative to the Western NGO’s evaluations and knowledge production, especially in respect to development and indigenous knowledge. By showing how community members are reframing the story of development, this paper demonstrates the usefulness of using participatory documentary photography in community-led evaluations, and helps balance the playing field by providing a much-needed alternative narrative of project evaluation.

Antigen Drives Regulatory B10 Cell Development and Function

B cells mediate immune responses via the secretion of antibody and interactions with other immune cell populations through antigen presentation, costimulation, and cytokine secretion. Although B cells are primarily believed to promote immune responses using the mechanisms described above, some unique regulatory B cell populations that negatively influence inflammation have also been described. Among these is a rare interleukin (IL)-10-producing B lymphocyte subset termed “B10 cells.” B cell-derived IL-10 can inhibit various arms of the immune system, including polarization of Th1/Th2 cell subsets, antigen presentation and cytokine production by monocytes and macrophages, and activation of regulatory T cells. Further studies in numerous autoimmune and inflammatory models of disease have confirmed the ability of B10 cells to negatively regulate inflammation in an IL-10-dependent manner. Although IL-10 is indispensable to the effector functions of B10 cells, how this specialized B cell population is selected in vivo to produce IL-10 is unknown. Some studies have demonstrated a link between B cell receptor (BCR)-derived signals and the acquisition of IL-10 competence. Additionally, whether antigen-BCR interactions are required for B cell IL-10 production during homeostasis as well as active immune responses is a matter of debate. Therefore, the goal of this thesis is to determine the importance of antigen-driven signals during B10 cell development in vivo and during B10 cell-mediated immunosuppression.

Chapter 3 of the dissertation explored the BCR repertoire of spleen and peritoneal cavity B10 cells using single-cell sequencing to lay the foundation for studies to understand the full range of antigens that may be involved in B10 cell selection. In both the spleen and peritoneal cavity B10 cells studied, BCR gene utilization was diverse, and the expressed BCR transcripts were largely unmutated. Thus, B10 cells are likely capable of responding to a wide range of foreign and self-antigens in vivo.

Studies in Chapter 4 determined the predominant antigens that drive B cell IL-10 secretion during homeostasis. A novel in vitro B cell expansion system was used to isolate B cells actively expressing IL-10 in vivo and probe the reactivities of their secreted monoclonal antibodies. B10 cells were found to produce polyreactive antibodies that bound multiple self-antigens. Therefore, in the absence of overarching active immune responses, B cell IL-10 is secreted following interactions with self-antigens.

Chapter 5 of this dissertation investigated whether foreign antigens are capable of driving B10 cell expansion and effector activity during an active immune response. In a model of contact-induced hypersensitivity, in vitro B cell expansion was again used to isolate antigen-specific B10 clones, which were required for optimal immunosuppression.

The studies described in this dissertation shed light on the relative contributions of BCR-derived signals during B10 cell development and effector function. Furthermore, these investigations demonstrate that B10 cells respond to both foreign and self-antigens, which has important implications for the potential manipulation of B10 cells for human therapy. Therefore, B10 cells represent a polyreactive B cell population that provides antigen-specific regulation of immune responses via the production of IL-10.

Statistical Dialog Management for Health Interventions

Research endeavors on spoken dialogue systems in the 1990s and 2000s have led to the deployment of commercial spoken dialogue systems (SDS) in microdomains such as customer service automation, reservation/booking and question answering systems. Recent research in SDS has been focused on the development of applications in different domains (e.g. virtual counseling, personal coaches, social companions) which requires more sophistication than the previous generation of commercial SDS. The focus of this research project is the delivery of behavior change interventions based on the brief intervention counseling style via spoken dialogue systems. Brief interventions (BI) are evidence-based, short, well structured, one-on-one counseling sessions. Many challenges are involved in delivering BIs to people in need, such as finding the time to administer them in busy doctors' offices, obtaining the extra training that helps staff become comfortable providing these interventions, and managing the cost of delivering the interventions. Fortunately, recent developments in spoken dialogue systems make the development of systems that can deliver brief interventions possible. The overall objective of this research is to develop a data-driven, adaptable dialogue system for brief interventions for problematic drinking behavior, based on reinforcement learning methods. The implications of this research project includes, but are not limited to, assessing the feasibility of delivering structured brief health interventions with a data-driven spoken dialogue system. Furthermore, while the experimental system focuses on harmful alcohol drinking as a target behavior in this project, the produced knowledge and experience may also lead to implementation of similarly structured health interventions and assessments other than the alcohol domain (e.g. obesity, drug use, lack of exercise), using statistical machine learning approaches. In addition to designing a dialog system, the semantic and emotional meanings of user utterances have high impact on interaction. To perform domain specific reasoning and recognize concepts in user utterances, a named-entity recognizer and an ontology are designed and evaluated. To understand affective information conveyed through text, lexicons and sentiment analysis module are developed and tested.

DEVELOPMENT OF A CONDENSED PHASE VELOCITY MAP IMAGING APPARATUS: DISSOCIATION PHOTODYNAMICS OF NITROGEN DIOXIDE AT 355 NM

The photochemistry of the polar regions of Earth, as well as the interstellar medium, is driven by the effect of ultraviolet radiation on ice surfaces and on the materials trapped within them. While the area of ice photochemistry is vast and much research has been completed, it has only recently been possible to study the dynamics of these processes on a microscopic level. One of the leading techniques for studying photoreaction dynamics is Velocity Map Imaging (VMI). This technique has been used extensively to study several types of reaction dynamics processes. Although the majority of these studies have utilized molecular beams as the main medium for reactants, new studies showed the versatility of the technique when applied to molecular dynamics of molecules adsorbed on metal surfaces. Herein the development of a velocity map imaging apparatus capable of studying the photochemistry of condensed phase materials is described. The apparatus is used to study of the photo-reactivity of NO2 condensed within argon matrices to illustrate its capabilities. A doped ice surface is formed by condensing Ar and NO2 gas onto a sapphire rod which is cooled using a helium compressor to 20 K. The matrix is irradiated using an Nd:YAG laser at 355 nm, and the resulting NO fragment is state-selectively ionized using an excimer-pumped dye laser. In all, we are able to detect transient photochemically generated species and can collect information on their quantum state and kinetic energy distribution. It is found that the REMPI spectra changes as different sections of the dissociating cloud are probed. The rotational and translational energy populations are found to be bimodal with a low temperature component roughly at the temperature of the matrix, and a second component with much higher temperature, the rotational temperature showing a possible population inversion, and the translational temperature of 100-200 K. The low temperature translational component is found to dominate at long delay times between dissociation and ionization, while at short time delays the high temperature component plays a larger role. The velocity map imaging technique allows for the detection of both the axial and radial components of the translational energy. The distribution of excess energy over the rotational, electronic and translational states of the NO photofragments provides evidence for collisional quenching of the fragments in the Ar-matrix prior to their desorption.

Karst hazard analysis towards the development of predictive karst models for southern Ontario

An investigation into karst hazard in southern Ontario has been undertaken with the intention of leading to the development of predictive karst models for this region. The reason these are not currently feasible is a lack of sufficient karst data, though this is not entirely due to the lack of karst features. Geophysical data was collected at Lake on the Mountain, Ontario as part of this karst investigation. This data was collected in order to validate the long-standing hypothesis that Lake on the Mountain was formed from a sinkhole collapse. Sub-bottom acoustic profiling data was collected in order to image the lake bottom sediments and bedrock. Vertical bedrock features interpreted as solutionally enlarged fractures were taken as evidence for karst processes on the lake bottom. Additionally, the bedrock topography shows a narrower and more elongated basin than was previously identified, and this also lies parallel to a mapped fault system in the area. This suggests that Lake on the Mountain was formed over a fault zone which also supports the sinkhole hypothesis as it would provide groundwater pathways for karst dissolution to occur. Previous sediment cores suggest that Lake on the Mountain would have formed at some point during the Wisconsinan glaciation with glacial meltwater and glacial loading as potential contributing factors to sinkhole development. A probabilistic karst model for the state of Kentucky, USA, has been generated using the Weights of Evidence method. This model is presented as an example of the predictive capabilities of these kind of data-driven modelling techniques and to show how such models could be applied to karst in Ontario. The model was able to classify 70% of the validation dataset correctly while minimizing false positive identifications. This is moderately successful and could stand to be improved. Finally, suggestions to improving the current karst model of southern Ontario are suggested with the goal of increasing investigation into karst in Ontario and streamlining the reporting system for sinkholes, caves, and other karst features so as to improve the current Ontario karst database.

A Social Networking Strategy for Improving Knowledge Management and Communication in the Travel Industry

In today’s technology-driven marketplace, the adoption and management of corporate and customer-facing Social Networking Sites (SNs) is often viewed as a key success factor for Travel Industry (TI) organisations. Knowledge management and the sharing of expertise and experiences through communication between internal and external stakeholders via social networks is an activity which TI organisations are aiming to exploit in order to improve the open sharing, retrieval, organisation and leveraging of knowledge. Through a study of currently-available literature relating to social networking adoption within the TI and a case study analysis of corporate social networking practices at three multi-national TI organisations (British Airways, Thomas Cook and Marriott Hotels), it may be observed that correlations exist between the development of social networking and the processes TI organisations now use to manage knowledge. We explore how these companies are currently utilizing SNs to improve knowledge management practices inside and outside of their organisational boundaries. From our analysis, lessons may emerge as to how TI companies are gaining competitive advantage through the use of social networking; a proposed strategy is identified to determine how TI organisations may make best use of social networks.