Representing Unevenly-Spaced Time Series Data for Visualization and Interactive Exploration

Aris, Aleks; Shneiderman, Ben; Plaisant, Catherine; Shmueli, Galit; Jank, Wolfgang

doi:10.1007/11555261_66

Cited by 42 publications

(27 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These efforts tend to focus on the challenge of adequately presenting high-volume data given such constraints as limited screen resolution, and on finding ways of dealing with the kinds of discontinuities that arise when values are sampled unevenly [5]. Another area of interest in visualising time series is interactivity.…”

Section: Visualisationmentioning

confidence: 99%

Automatic Generation of Textual Summaries from Neonatal Intensive Care Data

Portet

Reiter

Hunter

et al.

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

Section: Visualisationmentioning

confidence: 99%

Automatic Generation of Textual Summaries from Neonatal Intensive Care Data

Portet

Reiter

Hunter

et al.

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

“…Time series data which is sampled regularly but contains missing values can be seen as unregular sampled, or event-based data. Aris et al (Aris et al, 2005) developed for the TimeSearcher application four methods to deal with such unevenly spaced time series data; namely: sampled events, aggregated sampled events, event index and interleaved event index. Since neither method meets all the requirements of our casethe first two methods introduce new data at the sample points, the latter two only consider the order of events, not the specific time they appeared -we use a different method in handling missing values.…”

Section: Related Workmentioning

confidence: 99%

INTERACTIVE VISUAL ANALYSIS OF INTENSIVE CARE UNIT DATA - Relationship between Serum Sodium Concentration, its Rate of Change and Survival Outcome

Matković

Gan²,

Ammer

et al. 2012

Proceedings of the International Conference on Computer Graphics Theory and Applications

View full text Add to dashboard Cite

Abstract:In this paper we present a case study of interactive visual analysis and exploration of a large ICU data set. The data consists of patients' records containing scalar data representing various patients' parameters (e.g. gender, age, weight), and time series data describing logged parameters over time (e.g. heart rate, blood pressure). Due to the size and complexity of the data, coupled with limited time and resources, such ICU data is often not utilized to its full potential, although its analysis could contribute to a better understanding of physiological, pathological and therapeutic processes, and consequently lead to an improvement of medical care. During the exploration of this data we identified several analysis tasks and adapted and improved a coordinated multiple views system accordingly. Besides a curve view which also supports time series with gaps, we introduced a summary view which allows an easy comparison of subsets of the data and a box plot view in a coordinated multiple views setup. Furthermore, we introduced an inverse brush, a secondary brush which automatically selects non-brushed items, and updates itself accordingly when the original brush is modified. The case study describes how we used the system to analyze data from 1447 patients from the ICU at Guy's & St.Thomas' NHS Foundation Trust in London. We were interested in the relationship between serum sodium concentration, its rate of change and their effect on ICU mortality rates. The interactive visual analysis led us to findings which were fascinating for medical experts, and which would be very difficult to discover using conventional analysis methods usually applied in the medical field. The overall feedback from domain experts (coauthors of the paper) is very positive.

show abstract

“…Based on their previously developed TimeBoxes, which are rectangular direct-manipulation time series queries, they proposed an extension by introducing variable time timeboxes (VTT), permitting the specification of queries to allow uncertainty in the time axis. Four methods (sample events, aggregated sample events, event index and interleaved event index) representing the unevenly space time series data are studied in Aris et al [26]. Furthermore, TimeSearcher2 has been developed by Buono et al [27] combining both filter and pattern search capability in a new search interface.…”

Section: Introductionmentioning

confidence: 99%

An on-line SAX and HMM-based anomaly detection and visualization tool for early disturbance discovery in a dynamic industrial process

Yiakopoulos

Gryllias

Chioua³

et al. 2016

Journal of Process Control

View full text Add to dashboard Cite

a b s t r a c tIn order to achieve an optimum and successful operation of an industrial process, it is important firstly to detect upsets, equipment malfunctions or other abnormal events as early as possible and secondly to identify and remove the cause of those events. Univariate and multivariate statistical process control methods have been widely applied in process industries for early fault detection and localization.The primary objective of the proposed research is the design of an anomaly detection and visualization tool that is able to present to the shift operator -and to the various levels of plant operation and company management -an early, global, accurate and consolidated presentation of the operation of major subgroups or of the whole plant, aided by a graphical form.Piecewise Aggregate Approximation (PAA) and Symbolic Aggregate Approximation (SAX) are considered as two of the most popular representations for time series data mining, including clustering, classification, pattern discovery and visualization in time series datasets. However SAX is preferred since it is able to transform a time series into a set of discrete symbols, e.g. into alphabet letters, being thus far more appropriate for a graphical representation of the corresponding information, especially for the shift operator. The methods are applied on individual time records of each process variable, as well as on entire groups of time records of process variables in combination with Hidden Markov Models. In this way, the proposed visualization tool is not only associated with a process defect, but it allows also identifying which specific abnormal situation occurred and if this has also occurred in the past. Case studies based on the benchmark Tennessee Eastman process demonstrate the effectiveness of the proposed approach. The results indicate that the proposed visualization tool captures meaningful information hidden in the observations and shows superior monitoring performance.

show abstract

Representing Unevenly-Spaced Time Series Data for Visualization and Interactive Exploration

Cited by 42 publications

References 20 publications

Automatic Generation of Textual Summaries from Neonatal Intensive Care Data

Automatic Generation of Textual Summaries from Neonatal Intensive Care Data

INTERACTIVE VISUAL ANALYSIS OF INTENSIVE CARE UNIT DATA - Relationship between Serum Sodium Concentration, its Rate of Change and Survival Outcome

An on-line SAX and HMM-based anomaly detection and visualization tool for early disturbance discovery in a dynamic industrial process

Contact Info

Product

Resources

About