Designing Progressive and Interactive Analytics Processes for High-Dimensional Data Analysis

Abstract

In interactive data analysis processes, the dialogue between the human and the computer is the enabling mechanism that can lead to actionable observations about the phenomena being investigated. It is of paramount importance that this dialogue is not interrupted by slow computational mechanisms that do not consider any known temporal human-computer interaction characteristics that prioritize the perceptual and cognitive capabilities of the users. In cases where the analysis involves an integrated computational method, for instance to reduce the dimensionality of the data or to perform clustering, such non-optimal processes are often likely. To remedy this, progressive computations, where results are iteratively improved, are getting increasing interest in visual analytics. In this paper, we present techniques and design considerations to incorporate progressive methods within interactive analysis processes that involve high-dimensional data. We define methodologies to facilitate processes that adhere to the perceptual characteristics of users and describe how online algorithms can be incorporated within these. A set of design recommendations and according methods to support analysts in accomplishing high-dimensional data analysis tasks are then presented. Our arguments and decisions here are informed by observations gathered over a series of analysis sessions with analysts from finance. We document observations and recommendations from this study and present evidence on how our approach contribute to the efficiency and productivity of interactive visual analysis sessions involving high-dimensional data.

A Review and Characterization of Progressive Visual Analytics

“…• to realize responsive client-server visualizations using incremental data transmissions [4], • to make computational processes more transparent through execution feedback and control [5], • to steer visual presentations by prioritizing the display of regions of interest [6], • to provide fluid interaction by respecting human time constraints [7], or • to base early decisions on partial results, trading precision for speed [8].…”

Structuring visual exploratory analysis of skill demand

“…Exponential growth in data size and complexity and increased reliance on heterogeneous, third-party resources impact human ability to effectively extract, manage and make use of the knowledge content of the vast amounts of data that feed into task completion in today's data-rich and data-driven economy [16,[25][26][27]. Visual analytics provides a valuable tool that complements advanced human perception and analytical ability with automated data processing and mining.…”

“…Among others, they address the need for pre-processing due to the large number of dimensions and dataset size, to cluster data either along the dimensions defined or based on similarity in data point features. Liu et al, as do Hervás et al [5], Kerren et al [26], Turkay et al [27], highlight the benefits in functionality for (automated) animation and interactive data exploration, that gradually reveal ROIs and trends in data, aiding the construction of understanding of complex, static and dynamic datasets. Inselberg [30], Heer et al [25], among others, look at the benefits in multi-dimensional visualisation techniques such as scatterplot matrices and parallel coordinates.…”

“…Designing effective support for complex scenarios and data typically involves the exploration of the impact and use of multiple sub-components [4,12,26,27,29,32], following a process that may involve any or all of the following in an iterative cycle:…”

“…The need to manage increasingly large amounts of complex data is not an uncommon challenge in visual analytics [2,12,27]. However, screen real estate and resolution, processing power and human cognition place limits on how much data may be usefully displayed at once [1,28].…”