The development and preliminary testing of an instrument to measure the perceived economic burden of a subtype of arrhythmogenic right ventricular cardiomyopathy (ARVC) on patients and families in Newfoundland

The goal of this thesis was to develop, construct, and validate the Perceived Economic Burden scale to quantitatively measure the burden associated with a subtype Arrhythmogenic Right Ventricular Cardiomyopathy (ARVC) in families from the island of Newfoundland. An original 76 item self-administered survey was designed using content from existing literature as well as themes from qualitative research conducted by our team and distributed to individuals of families known to be at risk for the disease. A response rate of 37.2% (n = 64) was achieved between December 2013 and May 2014. Tests for data quality, Likert scale assumptions and scale reliability were conducted and provided preliminary evidence of the psychometric properties of the final constructed perceived economic burden of ARVC scale comprising 62 items in five sections. Findings indicated that being an affected male was a significant predictor of increased perceived economic burden in the majority of economic burden measures. Affected males also reported an increased likelihood of going on disability and difficulty obtaining insurance. Affected females also had an increased perceived financial burden. Preliminary results suggest that a perceived economic burden exists within the ARVC population in Newfoundland.

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.