Search
Navigator: RI | People | Steven F Roth
Graphics enhanced version of this site
Steven F Roth
Associate Research Professor
No longer a member of RI.
Jump to:
Biography |
Research interests |
Keywords |
Labs & Groups |
Projects |
Publications
Biography
Steven Roth, founder and former chief executive officer of Maya Viz Ltd., the South Side software design firm that was bought by a unit of General Dynamics in April, died in his sleep early Sunday morning, June 12, 2005. He was 53.
Obituary
Research interests
My research interests are in the general areas of artificial intelligence and cognitive psychology, with a focus on human-computer interaction and information visualization and exploration. Our students and staff are working on several projects within our research group and collaboratively on other projects in the School of Computer Science, providing HCI and related expertise as part of multidisciplinary teams. Our projects are addressing:
- the application of artificial intelligence to automate the creation of effective graphical presentations of large amounts of diverse information,
- novel techniques for human-computer interaction, including 3D visualization and direct manipulation; interactive, graphical data exploration environments for navigation, aggregation, and transformation of data into useful information,
- explanation generation using coordinated graphics and text produced by graphic and natural language systems, and
- interactive planning and scheduling environments for complex transportation and factory logistics systems.
SAGE: Automatic and interactive design of graphical presentations of information. The goal of this work is to provide computer systems with the ability to communicate with users through the automatic design of graphical and textual displays. The rationale is to give users access to graphic design expertise and to reduce the burden of assembling data, designing displays and interacting with complex interfaces to graphics packages. To accomplish this, we have been developing an application-independent graphic design system called SAGE. As input, SAGE takes data from applications or databases, and characterizations of the data and tasks to be performed. As output, it generates one or more displays which visually integrate the data in a way that supports the tasks.
Research issues include: representing the syntax and semantics of graphical conventions; encoding graphic selection, composition and other design knowledge; developing techniques for searching the space of alternative designs; representing the behavior of interactive graphical objects for manipulating data in displays, especially for exploring and aggregating large data sets; coordinating automatic and interactive design processes (i.e. coordinating human and machine expertise); coordinating the selection and integration of text and graphics to support explanation of results of planning and scheduling systems; developing approaches to dynamic redesign of displays to assimilate changing data.
Multi-media Explanation. We are developing ways to combine text generation with SAGE's graphics capabilities to produce coordinated explanations and summaries of large databases. This includes automatic selection of important content and expression of this content by distributing it between the two media. Explanations will be generated within an interactive data exploration environment so that both text and graphics serve as interfaces for exploring information in greater detail.
3D Visualization. We are developing techniques and systems and studying the HCI aspects of 3D visualization. The research is concerned with visualization coupled with new interactive techniques designed to enable performance of complex data manipulation and exploration tasks within a 3D environment. Research includes: creating new visualization metaphors for combinations of quantitative, geographic, temporal, categorical and hierarchical data; developing gesturing techniques for a variety of input devices (e.g. pointing, selecting, linking, directing movement) for providing user control in a 3D environment; assessing human perceptual ability for assimilating 3D representations.
Research interest keywords
computer graphics, data visualization, and human-computer interaction
Past Labs & Groups
Past Projects
-
AutoBrief - Generation of coordinated multimedia (natural language and information graphics) explanations from large and complex data sets.
-
Scheduling and Visualization - This project is investigating the development of next-generation environments for collaborative analysis and management of large-scale schedules.
-
Selective Dynamic Manipulation - A paradigm for interacting with visualizations that is based on the notion of physicalization, which uses the metaphor of creating ""physical"" objects to represent abstract data objects
-
System for Automated Graphics and Explanation - Mixed-initiative presentation system that supports visualization creation
-
Visage - Prototype software environment for exploring and visualizing large amounts of diverse information
-
Visual Query Environment - VQE provides direct-manipulation database queries for
operating on many objects at once.
Recent publications [View all 44 publications]
- Data Exploration across Temporal Contexts
M. Derthick and S.F. Roth
Proceedings of Intelligent User Interfaces (IUI '00), January, 2000.
[Abstract]
Download: pdf [192 KB] copyrighted
- Mapping Communicative Goals into Conceptual Tasks to Generate Graphics in Discourse
S. Kerpedjiev and S.F. Roth
Proceedings of Intelligent User Interfaces (IUI '00), January, 2000.
[Abstract]
Download: pdf [82 KB] copyrighted
- Efficient Multi-Object Dynamic Query Histograms
M. Derthick, J. Harrison, A. Moore, and S.F. Roth
Proceedings of the IEEE Information Visualization Conference (InfoVis '99), 1999.
[Abstract]
Download: pdf [284 KB] copyrighted
- Saying It In Graphics: from Intentions to Visualizations
S. Kerpedjiev, G. Carenini, N. Green, J. Moore, and S.F. Roth
Proceedings of the IEEE Symposium on Information
Visualization (InfoVis '98), October, 1998, pp. 97 - 101.
[Abstract]
Download: pdf [141 KB] copyrighted
- A Media-Independent Content Language for Integrated Text and Graphics Generation
N. Green, G. Carenini, S. Kerpedjiev, S.F. Roth, and J. Moore
Proceedings of the Workshop on Content Visualization and Intermedia Representations (CVIR'98) of the 17th International Conference on Computational Linguistics (COLING '98) and the 36th Annual Meeting of the Association for Computational Linguistics (ACL'98), August, 1998.
[Abstract]
Download: pdf [170 KB], ps.gz [49 KB] copyrighted
- Generating Visual Arguments: a Media-independent Approach
N. Green, S. Kerpedjiev, and S.F. Roth
AAAI98 Workshop on Representations for Multi-modal Human-Computer Interaction, July, 1998.
Download: pdf [107 KB], ps.gz [32 KB] copyrighted
- Describing Complex Charts in Natural Language: A Caption Generation System
V. Mittal, J. Moore, G. Carenini, and S.F. Roth
Computational Linguistics, Special issue on Natural Language
Generation, Vol. 24, No. 3, 1998, pp. 431 - 467.
[Abstract]
Download: pdf [382 KB] copyrighted
- An Interactive Visual Query Environment for Exploring Data
M. Derthick, J. Kolojejchick, and S.F. Roth
Proceedings of the ACM Symposium on User Interface Software and Technology, ACM Press, October, 1997, pp. 189-198.
Download: pdf [220 KB] copyrighted
- Coordinating declarative queries with a direct manipulation data exploration environment
M. Derthick, S.F. Roth, and J. Kolojejchick
Proceedings of the IEEE
Symposium on Information Visualization, October, 1997, pp. 65 - 72.
[Abstract]
Download: pdf [830 KB] copyrighted
- An Interactive Visualization Environment for Data Exploration
M. Derthick, J. Kolojejchick, and S.F. Roth
Proceedings of Knowledge Discovery in Databases, AAAI Press, August, 1997, pp. 2-9.
Download: pdf [227 KB] copyrighted
The Robotics Institute is part of the
School of Computer Science,
Carnegie Mellon University.
For updates and comments, please see these
instructions.
This page maintained by robotwebmaster@ri.cmu.edu