Authors

Supriya Garg

Type

Text

Type

Dissertation

Advisor

Mueller, Klaus | I. V. Ramakrishnan | Tamara L. Berg | Kevin T. McDonnell.

Date

2010-12-01

Keywords

Computer Science

Department

Department of Computer Science

Language

en_US

Source

This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.

Identifier

http://hdl.handle.net/11401/70992

Publisher

The Graduate School, Stony Brook University: Stony Brook, NY.

Format

application/pdf

Abstract

Visual Analytics is the integration of interactive visualization with analysis techniques to help answer questions, and find interesting"patterns" in a given dataset. This approach is useful in cases where human knowledge and intuition is required. Further, with the increasing size of data, an interactive visual interface can give the user control over what she wants to see. Towards this end, we develop a Visual Analytics (VA) infrastructure, rooted on techniques in machine learning and logic-based deductive reasoning. This system assists people in making sense of large, complex data sets by facilitating the generation and validation of models representing relationships in the data. The above data is often communicated among people, and by systems as graphs and other visual forms. However, currently the degree of semantics encoded in these representations is quite limited. In order to design more expressive icons, we present a framework that uses natural language text processing and global image databases to help users identify metaphors suitable to visually encode abstract semantic concepts. While using images as forms of communication, people often like to add annotations in forms of text, arrows, etc. to make them more expressive. The process of selecting a color for the annotation can often be difficult if the background has a variation in color and texture. Our tool Magic marker helps users by suggesting appropriate colors for annotating an image. All colors in a modified CIE-Lab color-space are assigned a preference value, and users can navigate this space using an interactive interface. This interactivity also allows them to make the final choice based on personal preferences.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.