TU Darmstadt / ULB / TUbiblio

Visual Analysis of Relations in Attributed Time-Series Data

Steiger, Martin and Bernard, Jürgen and Schader, Philipp and Kohlhammer, Jörn (2015):
Visual Analysis of Relations in Attributed Time-Series Data.
Eurographics Association, Goslar, In: EuroVA 2015, DOI: 10.2312/eurova.20151105,
[Conference or Workshop Item]

Abstract

In this paper, we present visual-interactive techniques for revealing relations between two co-existing multivariate feature spaces. Such data is generated, for example, by sensor networks characterized by a set of (categorical) attributes which continuously measure physical quantities over time. A challenging analysis task is the seeking for interesting relations between the time-oriented data and the sensor attributes. Our approach uses visualinteractive analysis to enable analysts to identify correlations between similar time series and similar attributes of the data. It is based on a combination of machine-based encoding of this information in position and color and the human ability to recognize cohesive structures and patterns. In our figures, we illustrate how analysts can identify similarities and anomalies between time series and categorical attributes of metering devices and sensors.

Item Type: Conference or Workshop Item
Erschienen: 2015
Creators: Steiger, Martin and Bernard, Jürgen and Schader, Philipp and Kohlhammer, Jörn
Title: Visual Analysis of Relations in Attributed Time-Series Data
Language: English
Abstract:

In this paper, we present visual-interactive techniques for revealing relations between two co-existing multivariate feature spaces. Such data is generated, for example, by sensor networks characterized by a set of (categorical) attributes which continuously measure physical quantities over time. A challenging analysis task is the seeking for interesting relations between the time-oriented data and the sensor attributes. Our approach uses visualinteractive analysis to enable analysts to identify correlations between similar time series and similar attributes of the data. It is based on a combination of machine-based encoding of this information in position and color and the human ability to recognize cohesive structures and patterns. In our figures, we illustrate how analysts can identify similarities and anomalies between time series and categorical attributes of metering devices and sensors.

Publisher: Eurographics Association, Goslar
Uncontrolled Keywords: Business Field: Visual decision support, Research Area: Human computer interaction (HCI), Time series analysis, Human-computer interaction (HCI), User-centered design
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Interactive Graphics Systems
Event Title: EuroVA 2015
Date Deposited: 08 May 2019 07:09
DOI: 10.2312/eurova.20151105
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item