A Visual Analytics Approach to High-Dimensional Logistic Regression Modeling and its Application to an Environmental Health Study
In the domain of epidemiology, logistic regression modeling is widely used to explain the relationships among explanatory variables and dichotomous outcome variables. However, logistic regression modeling faces challenges such as overfitting, confounding, and multicollinearity when there is a large number of explanatory variables. For example, in the birth defect study presented in this paper, variable selection for building high quality models to identify risk factors from hundreds of pollutant variables is difficult. To address this problem, we propose a novel visual analytics approach to logistic regression modeling for high-dimensional datasets. It leverages the traditional modeling pipeline by providing (1) intuitive visualizations for inspecting statistical indicators and the relationships among the variables and (2) a seamless, effective dimension reduction pipeline for selecting variables for inclusion in high quality logistic regression models. A fully working prototype of this approach has been developed and successfully applied to the birth defect study, which illustrates its effectiveness and efficiency.
Spot-Tracking Lens: A Novel Interaction for Animated Bubble Charts
Animated bubble charts are fun for visualizing temporal data, but their effectiveness in analysis has been questioned in recent studies. In this paper, we discover the power of animated bubble charts in revealing local, short-lived patterns and maximize this power using a novel spot-tracking lens technique. The basic idea is to design the animated visualization as what can be seen from a zoom camera lens over a focus object. The visualization enlarges local context of the object and automatically tracks its motion with a spotlight during the animation. This spot-tracking scheme utilizes selective visibility with moving focus to allow observers to concentrate on the tracked object and its changing context. The spot-tracking lens can effectively reduce change blindness and enable users to discover volatile local patterns around the tracked object. Those patterns are difficult to capture in static approaches such as small multiples and traces. Our new technique is useful in egocentric visual explorations where users gain insights on unfamiliar objects when observing their relationships with a familiar object. In progressive visual explorations, it can facilitate users in conducting detailed analysis on focal objects and their contexts and thus generate new hypotheses based on the observations. We explore the design space of camera movement, highlighting, labeling, and play control. Several user studies have been conducted to evaluate the effectiveness and efficiency of the new technique.
