The analysis of data streams has become quite important in recent years, and is being studied intensively in fields such as database management and data mining. Although some researchers in data and information visualization have investigated the visual analytics of streaming data to a certain degree, there are some obvious limitations in existing work: (1) a lack of effective techniques to show how data patterns change over time; and (2) limited ability to represent multivariate correlations. In this paper, we propose a framework to visualize multivariate data streams via a combination of windowing and sampling strategies. In order to help users observe how data patterns change over time, we display not only the current sliding window but also abstractions of past data in which users are interested. Sampling is applied within each single sliding window to help reduce visual clutter as well as preserve data patterns. Further, we allow different windows to have different sampling ratios to reflect how interested the user is in the contents. We use a DOI (degree of interest) function to represent users’ interest in the data within a set of windows. Users can apply two types of pre-defined DOI functions. An interactive tool also allows users to adjust the DOI function online, in a manner similar to transfer functions in volume visualization, to enable a trial-and-error exploration process. In order to visually convey the change of multidimensional correlations, we designed four layout strategies. User studies showed that three of these are effective techniques to achieve the above goal compared to traditional time-series data visualization techniques. Based on this evaluation experiment, we derived a guide to advise data analysts and visualization system developers on how to choose appropriate layout strategies in terms of the characteristics of datasets and data analysis tasks. Case studies are discussed to show the effectiveness of DOI functions and the various visualization techniques.
, Ward, Matthew
, Rundensteiner, Elke A.
(2010). Visual Analysis of Multivariate Data Streams Based on DOI Functions. .
Retrieved from: https://digitalcommons.wpi.edu/computerscience-pubs/18