Master of Education (MEd), Bowling Green State University, 2011, Career and Technology Education/Technology

As the need for educating traditional and non-traditional students increases and budgets decrease, the demand for higher education institutions to implement creative ways to provide effective customer service to students has never been more critical. This research studied the potential implementation of an Agent for Interactive Student Assistance (AISA) application in Bowling Green State University's (BGSU's) College of Technology and its impact on student engagement and recruitment. AISA is defined as an interactive, human-like, avatar-based online student assistance application with voice and text recognition that provides answers to students' administrative-related most frequently asked questions. The avatar-based application would provide cognitive responses using voice and non-verbal communication with a 90% accuracy rate. BGSU College of Technology undergraduate and graduate students during the 2009/2010 and 2010/2011 academic years were the population of this study consisting of 940 students. The approach of this study was quantitative, post positivist with an expected outcome in the form of an alternate hypothesis tested against a null hypothesis. One survey was administered to the population with a response rate of 9%. Favorable results were found with 91% of students indicating they would or may use an AISA application if provided the opportunity. One proportion z tests showed that, overall, students would not experience a negative impact on engagement and BGSU's College of Technology would not experience a decrease in new students.

Committee: Terry Herman PhD (Committee Chair); Gary Benjamin PhD (Committee Member); Anthony Fontana (Committee Member) Subjects: Education; Educational Technology

Master of Science (MS), Wright State University, 2022, Computer Science

Promoting awareness of social determinants of health (SDoH) among healthcare providers is important to improve the patient care experience and outcome as it helps providers understand their patients in a better way which can facilitate more efficient and effective communication about health conditions. Healthcare professionals are typically educated about SDoH through lectures, questionaries, or role-play-based approaches; but in today's world, it is becoming increasingly possible to leverage modern technology to create more impactful and accessible tools for SDoH education. Wright LIFE (Lifelike Immersion for Equity) is a simulation-based training tool especially created for this purpose. It is a mobile app that would be available on both Google Play and Apple Store for easy access to the providers. This highly realistic, interactive, and captivating app is essential for creating mindfulness about SDoH and generating long-lasting compassion and empathy in health care workers for their real patients and helping them to build a good clinician-patient relationship. An important aspect of this simulation is the realism of the characters and their behavior. This thesis specifically focuses on the strategy and experience of designing and developing realistic human character models and animations so that the players connect naturally and deeply with the virtual characters. This contributes to the generation of a greater level of empathy in the providers and decreases the level of biases. In addition to its contribution to creating efficient design methodologies, this effort also resulted in a portfolio of high-quality, low-memory multi-modal avatars resembling diverse people of various ethnicities, ages, body types, and gender.

Committee: Yong Pei Ph.D. (Advisor); Paul J. Hershberger Ph.D. (Committee Member); Thomas Wischgoll Ph.D. (Committee Member) Subjects: Computer Science; Design; Educational Software; Health Care; Information Technology; Systems Design

Master of Science, The Ohio State University, 2008, Computer Science and Engineering

This work conveys the personality of a virtual character through its posture. A standard personality model from literature is integrated with a customized behavior model to design a system that takes the high-level personality configuration of the character as input. Several works from the field of nonverbal behavior have been surveyed and a set of rules that map psychological states to physiological behavior have been formulated. These rules have been implemented in a system that outputs postures given high level psychological states or personality configurations as inputs.

Committee: Richard Parent Phd (Advisor); Matthew Lewis Phd (Committee Member) Subjects: Computer Science