Sinjini Mitra

Multimodal biometrics for enhanced mobile device security

Fusing information from multiple biometric traits enhances authentication in mobile devices.

Exploring Social Media for Health and Wellness: A Health Plan Case Study

The use of social media for health and wellness promotion is a relatively new concept. Nonetheless, several early adopting health plans and provider organizations have begun to design and pilot social and mobile media platforms to empower members to enhance self management of health and wellness goals. In this case study of a large health plan in Pennsylvania, the authors describe the design and execution of a member survey to identify some factors that are significantly associated with interest in adopting such technology platforms for obtaining health-related information and services. Analysis of relevant data from more than 4,000 responses indicates significant differences among important segments of the member population defined with respect to demographic factors, level of computer and social media usage, and frequency of engagement in specific online activities. They anticipate that these insights can assist health plans to develop and deploy targeted services and tools for health and wellness management.

Engagement with Social Media Platforms via Mobile Apps for Improving Quality of Personal Health Management: A Healthcare Analytics Case Study

Patient engagement in self health and wellness management has been identified as an important goal in improving health outcomes. As a result, the use of mobile and social media for health and wellness promotion is gathering considerable momentum. Several early adopting health plans and provider organizations have begun to design and pilot social and mobile media platforms to empower members to enhance self management of health and wellness goals. Based on a member survey of a large health plan in Pennsylvania, the authors identify factors that are significantly associated with member interest in adopting such technology platforms for obtaining health related information and services. Analysis of relevant data from more than 4,000 responses from health plan members indicate significant effects of several factors such as age, gender, general health condition including presence of chronic conditions like diabetes and high blood pressure, level of computer and social media usage and frequency of engaging in different online activities such as banking, shopping, and emailing. This analysis allows us to identify important consumer segments that are correlated with professed willingness to use applications and programs offered by the health plan. Besides, the authors also develop statistical models to predict peoples odds of adopting health-related mobile apps and identify the significant predictors thereof. The authors anticipate that these insights can assist health plans to develop and deploy targeted services and tools through integration of mobile and social media platforms for health and wellness management.

Inference for performance evaluation of fingerprint identification systems based on a hierarchical random effects model

ABSTRACT Biometric authentication is widely employed today, from immigration and border control to ensuring security on mobile devices. Just as it is important to devise efficient biometric systems, it is equally or even more important to evaluate their performance. This article introduces a novel method of performance evaluation, particularly to assess the “scalability” of fingerprint identification systems using a Bayesian inference technique based on hierarchical random effects models. We also extend this model to predict false alarm probabilities on “watch-lists.” Both of these aspects are important for national security applications of such systems. We illustrate our approach using three fingerprint databases.

Community Issues in American Metropolitan Cities: A Data Mining Case Study

The city of San Francisco in California has 826,000 residents and is growing slowly compared to other large cities in the western United States, facing concerns such as an aging population and flight of families to nearby suburbs. This case study investigates the social and demographic factors that are causing this phenomenon based on data that were collected by San Franciscos city controllers office in its annual survey to residents. By using data analytics, we can predict which residents are likely to move away, and this help us infer which factors of city life and city services contribute to a residents decision to leave the city. Results of this research indicate that factors like public transportation services, public schools, and personal finances are significant in this regard, which can potentially help the city of San Francisco to prioritize its resources in order to better retain its locals.

Motivational Impacts on Intent to Use Health-Related Social Media

ABSTRACT Today, the healthcare industry is challenged with promoting health and disease prevention through various customer outreach initiatives. Social media tools enable customers to aquire health information without the need to visit their doctor. Yet, healthcare providers do not understand what motivates customers to use such technology for health-related purposes. In this study, an online survey was conducted with 4,058 participants. Using SEM techniques, the results indicate that previous experience with online health-related searches serves as a direct and indirect motivational driver. While the direct relationship between prior experience with online health-related searches can be mostly positive, it can be weakened and impacted by inhibitors that create online use concerns (e.g., privacy and confidentiality). Furthermore, health condition may determine the level of interest people may have for health-related social media sites as well. This research provides a deeper understanding about motivational factors that impact the intent to use health-related social media.

Impact of Supplemental Instruction on Business Courses: A Statistical Study

Many students in quantitative business courses are struggling. One technique designed to support such students is Supplemental Instruction (SI), which is most popular in the science, technology, engineering, and mathematics (STEM) disciplines. In this paper, we show the positive impact of SI on student performance in two bottleneck business courses in a large university. Our evaluation results establish that (i) SI has a statistically significant effect on students’ likelihood of passing both courses (after controlling for background variables), (ii) SI is more helpful for students identified as at risk than for those who are not, and (iii) it is important to consistently attend SI sessions for greater success. We also present models to predict consistent student attendance based on background factors with 90% accuracy and conclude with a brief qualitative study about students’ self-perception of SI and the professional development attained by SI leaders.

Hidden Markov Models for feature-level fusion of biometrics on mobile devices

Although biometrics have forayed into the mobile world, most current approaches rely on a single biometric modality. This limits their recognition accuracy in uncontrolled conditions. For example, performance of face and voice recognition systems may suffer in poorly lit and noisy settings, respectively. Integration of identifying information from multiple biometric modalities can help solve this problem; high-quality identifying information in one modality can compensate for the absence of such information in a modality affected by uncontrolled conditions. In this paper, we present a novel multimodal biometric scheme that uses Hidden Markov Models to consolidate data from face and voice biometrics at the feature level. An implementation on the Samsung Galaxy S5 (SG5) phone using a dataset of face and voice samples captured using SG5 in real-world operating conditions, yielded 4.18% and 9.71% higher recognition accuracy than face and voice single-modality systems, respectively.

Biometrics in a Data Driven World: Trends, Technologies, and Challenges

Biometrics in a Data Driven World: Trends, Technologies, and Challenges aims to inform readers about the modern applications of biometrics in the context of a data-driven society, to familiarize them with the rich history of biometrics, and to provide them with a glimpse into the future of biometrics. The first section of the book discusses the fundamentals of biometrics and provides an overview of common biometric modalities, namely face, fingerprints, iris, and voice. It also discusses the history of the field, and provides an overview of emerging trends and opportunities. The second section of the book introduces readers to a wide range of biometric applications. The next part of the book is dedicated to the discussion of case studies of biometric modalities currently used on mobile applications. As smartphones and tablet computers are rapidly becoming the dominant consumer computer platforms, biometrics-based authentication is emerging as an integral part of protecting mobile devices against unauthorized access, while enabling new and highly popular applications, such as secure online payment authorization. The book concludes with a discussion of future trends and opportunities in the field of biometrics, which will pave the way for advancing research in the area of biometrics, and for the deployment of biometric technologies in real-world applications. The book is designed for individuals interested in exploring the contemporary applications of biometrics, from students to researchers and practitioners working in this field. Both undergraduate and graduate students enrolled in college-level security courses will also find this book to be an especially useful companion.

The Nature, Antecedents, and Impacts of Visuo-Spatial Mental Models of Web Interface Design

This paper develops an integrated theoretical framework that links visuo-spatial mental models of Web interface design to user satisfaction, and reports on a controlled experiment aimed at investigating the nature of these mental models. Using data from more than 500 subjects in conjunction with both graphical and statistical analyses, we find that Web users possess a strongly cohesive shared mental model of the way in which a Web interface should be designed. In addition to describing and quantifying this shared visuo-spatial mental model, this paper also shows how both experience and the physiological properties of the human visual system give rise to such models, and discusses the implications of the results for organizational website design, scientific theory, and future research in this area.