Visual Analytics State of the Art
Universitetet i Bergen
Abstract Visual analytics is a ﬁeld within Visualisation, which tries to deal with analysis of huge data. Visual analytics is more than just Visualisation. It is more an integration of data analysis, human innate abilities andvisualisation. [KMSZ06] According to J. Thomas and K. Cook [TC05] Visual analytics is the science of analytical reasoning supported by interactive visual interfaces. This is to say that it should facilitate humans analytical abilities, and hence give us the ability of extracting relevant data from sometimes huge datasets. This information is then visualised to the user in an appropriate way , while alsoallowing for interactive control of the process from the underlying data trough the algorithms. In this article I will try to look at the current state of Visual Analytics, with a focus on recent advancements. Categories and Subject Descriptors (according to ACM CCS): I.3.6 [Computer Graphics]: Methodology and Techniques
1. Introduction The ﬁeld of Visual Analytics has grown out of the ﬁelds ofinformation and scientiﬁc visualization. [KMSZ06]. It started in the United States from an initiative by the Department of Homeland Security. In response to the september 11. events, and in an effort to prevent future attacks, the National Institute for Visual Analytics was created. They were to develop a means to analyse the vast amounts of information available to the variuos governmentalinstitutions in the United States. The primary focus here was to detect and analyse potential threaths to the United States, such as terrorist attacks, in advance so that they can hopefully be prevented. Theese new techniques and priciples are what is meant by Visual Analytics. According to [TC05], you need human judgement to make sense of and detect important and relevant informations and patterns inincomplete, inconsistent, and potentially deceptive information which in addition could be rapidly changing. They further state that Visual analytics is the science of analytical reasoning facilitated by inter-active visual interfaces. Visual analytics can also be a means to detect the expected and discover the unexpected [TC05], arriving at meaningful conclusions in reasonable time and alsocommunicate these results to third parties.
c Morten Bendiksen.
1.1. Related research There are several related ﬁelds of research which can be identiﬁed. See the ﬁgure 1 below.
Figure 1: This picture gives an overview of different diciplines and problem areas that relates to visual analytics. Image courtesy of [KMSZ06]
As we see Visual analytics is more than just visualization and can rather beseen as an integrated approach combining visual- ization [KMSZ06], human factors and data analysis. Figure 1 shows many different ﬁelds relating to visual analytics. Within the visualization, there is use of information analytics, geospatial analytics, and scientiﬁc analytics. The
Morten Bendiksen / Visual Analytics State of the Art
human factors (e.g., inter- action, cognition, perception,collaboration, presentation, and dissemination) is important in the collaboration between human and computer and in the making of decisions. We want to create results on the basis of the analytical efforts. then we want to present this information so that it is easily understandable for the target audience. On the side of data analysis we make use of techniques from the ﬁelds of data managementknowledge representation, knowledge discovery, and statistical analytics. According to [KMSZ06] visual analytics is not likely to become a separate ﬁeld of study, but its inﬂuence will spread over the researchareas it comprises [KMSZ06] presents the following mantra for visual analytics: “Analyse First Show the Important Zoom, Filter and Analyse Further Details on Demand” Visual analytics combine...