Visual exploration of frequent patterns in multivariate time series

Lammarsch²,

Computer Graphics Forum

et al. 2018

To complement the currently existing definitions and conceptual frameworks of visual analytics, which focus mainly on activities performed by analysts and types of techniques they use, we attempt to define the expected results of these activities. We argue that the main goal of doing visual analytics is to build a mental and/or formal model of a certain piece of reality reflected in data. The purpose of the model may be to understand, to forecast or to control this piece of reality. Based on this model‐building perspective, we propose a detailed conceptual framework in which the visual analytics process is considered as a goal‐oriented workflow producing a model as a result. We demonstrate how this framework can be used for performing an analytical survey of the visual analytics research field and identifying the directions and areas where further research is needed.

Section: Surveying the Visual Analytics Researchmentioning

confidence: 99%

Viewing Visual Analytics as Model Building

Lammarsch²,

Computer Graphics Forum

et al. 2018

2016 IEEE Conference on Visual Analytics Science and Technology (VAST)

“…SAX achieves dimensionality reduction by segmenting the data based on a user specified segment length, assigning a symbol to each segment, and uses a sliding window of user specified length to generate a symbolic approximation. Different methods have tried to visualize the symbolically approximated time series data using bitmaps [26], VizTree based representations [33] and coloured rectangles [17] that allow users to visually explore the data. The SAX algorithm uses piecewise aggregate approximation (PAA) [23] for dimensionality reduction, but this method cannot capture local trends, and in turn the overall general shape of the time series, due to smoothing of perceptually important points (PIP) [13].…”

Section: Dimensionality Reductionmentioning

confidence: 99%

Shape grammar extraction for efficient query-by-sketch pattern matching in long time series

Muthumanickam

Vrotsou

Cooper

et al. 2016

Figure 1: A schematic representation of the work flow. A sketched pattern of interest is matched to the time-series data. Efficient approximation, classification and symbol assignment, based on gradient ratios, enables real-time pattern searching within very large time-series. The steps depicted with green boxes are executed when a new input time series is loaded or when it undergoes hierarchical approximation, yellow boxes only when a new sketch is entered. ABSTRACTLong time-series, involving thousands or even millions of time steps, are common in many application domains but remain very difficult to explore interactively. Often the analytical task in such data is to identify specific patterns, but this is a very complex and computationally difficult problem and so focusing the search in order to only identify interesting patterns is a common solution. We propose an efficient method for exploring user-sketched patterns, incorporating the domain expert's knowledge, in time series data through a shape grammar based approach. The shape grammar is extracted from the time series by considering the data as a combination of basic elementary shapes positioned across different amplitudes. We represent these basic shapes using a ratio value, perform binning on ratio values and apply a symbolic approximation. Our proposed method for pattern matching is amplitude-, scale-and translation-invariant and, since the pattern search and pattern constraint relaxation happen at the symbolic level, is very efficient permitting its use in a real-time/online system. We demonstrate the effectiveness of our method in a case study on stock market data although it is applicable to any numeric time series data.

“…Such is a common action in many productivity CASTs such as Microsoft Office. Hao et al (2012) discussed an example of an implementation of blending in the context of visual analytics. They introduced an interaction technique called ‗motif merging', which facilitates the exploration of frequently occurring patterns in time-series datasets.…”

Section: Blendingmentioning

confidence: 99%

Interaction Design for Complex Cognitive Activities with Visual Representations: A Pattern-Based Approach

Sedig

Parsons²

2013

THCI

109

156

This paper is concerned with interaction design for visualization-based computational tools that support the performance of complex cognitive activities, such as analytical reasoning, sense making, decision making, problem solving, learning, planning, and knowledge discovery. In this paper, a number of foundational concepts related to interaction and complex cognitive activities are syncretized into a coherent theoretical framework. This framework is general, in the sense that it is applicable to all technologies, platforms, tools, users, activities, and visual representations. Included in the framework is a catalog of 32 fundamental epistemic action patterns, with each action pattern being characterized and examined in terms of its utility in supporting different complex cognitive activities. This catalog of action patterns is comprehensive, covering a broad range of interactions that are performed by a diverse group of users for all kinds of tasks and activities. The presented framework is also generative, in that it can stimulate creativity and innovation in research and design for a number of domains and disciplines, including data and information visualization, visual analytics, digital libraries, health informatics, learning sciences and technologies, personal information management, decision support, information systems, and knowledge management.