Visual Exploration of Time-Series Data

full screen picture with link Pattern Search

TimeSearcher 1

TimeSearcher 2

Graph Envelope Magazine Example
Data/Query Envelope
Download TimeSearcher 1
Download TimeSearcher 2


TimeSearcher 3

TimeSearcher 3 - Forecasting Wizard




TimeSearcher 1

Project Description:

Widespread interest in discovering features and trends in time- series has generated a need for tools that support interactive exploration. We have built a prototype environment for interactive querying and exploration of time-series data. Queries are built using timeboxes: a powerful graphical, direct-manipulation metaphor for the specification of queries over time-series datasets. These timeboxes support interactive formulation and modification of queries, thus speeding the process of exploring time-series data sets and guiding data mining. The prototype includes windows for timebox queries, individual time-series, and details-on-demand. Other features include drag-and-drop support for  query-by-example and graphical envelopes for displaying the extent of the entire data set and result set from a given query.

The time-series prototype is built using HCIL's Piccolo Toolkit for Zoomable User Interfaces.

Participants:

Harry Hochheiser, Graduate Student Computer Science
Ben Shneiderman, Professor Computer Science

Papers:

Availability & Download

TimeSearcher can be downloaded for academic and non-commercial use.

If you are interested would like to use TimeSearcher for corporate use, please contact the University of Maryland, Office of Technology Commercialization. We are also willing to make source code available to collaborators: please contact hsh@cs.umd.edu for details.

For users who have a copy of TimeSearcher, a User Manual is available.

Press:

Related articles:

Support:

This research has been partially supported by funding from an AOL Fellowship to Harry Hochheiser, 2000-2001.


TimeSearcher 2

Project Description:

TimeSearcher 2 extends the research efforts of TimeSearcher 1, by visualizing long time series (>10,000 time points) and providing an overview that allows users to zoom into areas of interest. It provides the ability to view multivariate time series data, by showing up to ten simultaneous plots on the same screen. TimeSearcher 2 continues to have the TimeBox widget, but certain features of TimeSearcher 1, such as angular query, leaders and laggards have yet to be implemented. TimeSearcher 2 has been completely re-written in C# using the .NET framework. A powerful addition is the SearchBox that enables users to select a subset of an existing time series and search for similar patterns throughout the data. Users have four controls over the acceptable similarities. Our current efforts are focused on visualizing unevenly-spaced time series data, such as found in eBay auctions.

Current Participants:

Aleks Aris Graduate Student Computer Science
Ben Shneiderman Professor Computer Science
Catherine Plaisant Associate Research Scientist UMIACS
Galit Shmueli Assistant Professor Decision and Information Technologies Department
Wolfgang Jank Assistant Professor Decision and Information Technologies Department

Previous Participants:

Paolo Buono Assistant Professor Department of Informatics (U. of Bari, Italy)
Amir Khella Graduate Student Computer Science

Papers:

Availability & Download

TimeSearcher 2 can be downloaded for academic and non-commercial use.

TimeSearcher 2 is available for educational and research purposes. If you are interested to use TimeSearcher for corporate internal distribution or for inclusion in products, please contact Ben Shneiderman and Catherine Plaisant .

For users who have a copy of TimeSearcher 2, a User Manual is available.

Press:

Related Papers, Reports, and Examples:

Video:

Current Support:

This research has been supported by NSF grant DMI-0205489, the Center for Electronic Markets and Enterprises, the Smith Technology Integration Initiative, and the R.H. Smith School of Business at the University of Maryland.

Previous Support:

We'd like to thank to ChevronTexaco for the support they provided for our ongoing research efforts.

TimeSearcher 3

Project Description:

Time-series forecasting has a large number of applications. Users with a partial time series for auctions, new stock offerings, or industrial processes desire estimates of the future behavior. We present a data driven forecasting method and interface called Similarity-Based Forecasting (SBF). A pattern matching search in a dataset of historical time series produces a subset of curves similar to the partial time series. The forecast is displayed graphically as a river plot showing statistical information about the SBF subset. A forecasting preview interface allows users to interactively explore alternative pattern matching parameters and see multiple forecasts simultaneously. User testing with 8 users demonstrated advantages and led to improvements.

Participants:

Paolo Buono Assistant Professor Department of Informatics (U. of Bari, Italy)
Catherine Plaisant Associate Research Scientist UMIACS
Adalberto Simeone Graduate Student Department of Informatics (U. of Bari, Italy)
Aleks Aris Graduate Student Computer Science
Ben Shneiderman Professor Computer Science
Galit Shmueli Assistant Professor Decision and Information Technologies Department
Wolfgang Jank Assistant Professor Decision and Information Technologies Department

Papers:

Availability & Download

TimeSearcher 3 is currently not available for download publicly.
For private inquiries, contact
Ben Shneiderman and Catherine Plaisant.

Related Project:


TimeSearcher 1 last updated: 3 October 2002
This page last updated: