Effects of User, System & Data characteristics on the Visual Exploration of Time-Series data
Effects of User, System & Data characteristics on the Visual Exploration of Time-Series data
Salomon Eisler, PhD. student at the department of Industrial Engineering
Advisor:Prof. Joachim Meyer
Abstract:
Interactive visualizations play a crucial role in modern data analysis and decision-making. These processes inherently involve human interaction—both with the data and with its visual representations. Our research develops information-theoretic models to better understand these interactive dynamics, treating visualizations as information sources and users as virtual information channels. We validate our models through empirical experiments, taking into account individual user characteristics, and further interpret the findings through the lenses of decision theory, psychophysics, and behavioral judgment and decision-making frameworks. While information theory has previously been applied to study interactive visualizations, prior work has largely overlooked the human information-processing aspect. Our approach bridges this gap by focusing on the human component, emphasizing how user traits, dataset properties, and software design collectively influence interaction, analysis outcomes, and decisions. To validate our models, we conducted experiments involving a fundamental task: user classification of patterns in time-series line plots. These simple, intuitive visualizations are both widely used and “information lossless,” making them ideal for isolating and studying core aspects of human-data interaction. Insights from this research lay the groundwork for applying our models to more complex visualizations and analytical scenarios.
Bio:
Salomon Eisler is a PhD candidate in the Department of Industrial Engineering at Tel Aviv University,He holds an M.Sc. in Electrical Engineering from the Technion - Israel Institute of Technology. His research interests are in Visualizations, Information Theory, and Individual Differences.
