Project Teams List

Group 1:

Title: The UAF Visualization Tool

Members: Jihane Najdi, Amine Chigani, Chris Catanzaro

Link: http://csgrad.cs.vt.edu/~ccatanza/InfoVis.htm

Abstract: Our project is to develop a usability engineering tool based on the User Action Framework. This tool is motivated by the variability of analysis processes and facilitates the analysis of existing problem data with the goal of process improvement. Usability engineers spend considerable resources performing usability testing and analyzing the resulting data, but their work is often limited to one development effort or a small set of similar efforts. To extend the analysis process to multiple diverse efforts, it is necessary to categorize and store data in a consistent manner and have techniques for discovering patterns in that data. Our tool is designed to work with problem data that has been organized according to a hierarchical framework of usability concepts, which ensures consistency through completeness and precision. In addition, our tool helps engineers discover weaknesses in their process through exploratory browsing of a visualization of the tree structure and visual filtering based on cost, criticality, and keywords.

      Currently, we have a functional prototype that we will base our future research and implementations. The functional prototype has been proven useful thus far. However, we feel more steps have to be taken to enhance the current functionality as well as develop more features as research uncovers more needs and wants for the tool.

Some interesting visualization obstacles that are present in this visualization task are that of illustrating each node possibly containing multiple problems and also showing many-to-many mapping of problems to keywords.

      This tool gives a usability expert or project manager the functionality to visualize the problem areas. Also the tool gives the ability to sort, search, compare, and query for problems by cost, criticality or keywords.

      In order to progress in the development of the UAF Visualization tool we will have a preliminary meeting with Dr. Hartson to discuss a usability practitioner survey that will be sent out in order to gather feed-back from current practitioners. We will also meet to hear his thoughts about the tool as he has mentioned during his research with the UAF he has thought of useful items to visualize.

      From this meeting we will then set milestones for the project in order to fulfill the project goals.

Group 2:

Title: Information-Rich Virtual Environments (IRVE)

Members: Vladimir Glina, Lauren Shupp, James Volpe

Abstract: The IRVE project is an evaluation of strategies for visualizing data in virtual environments (VE). Methods such as interactively linking VEs to information vsualizations and visually embedding abstract data ithin the VE are explored. The project compares the effectiveness of these interactive strategies.

Group 3:

Title: Training Wheels

Members: Pradyut Bafna, Matthew Phillips, Johnny L Sam Rajkumar

Link: netpire.com/school/infovis/index.htm

Abstract: The goal of the training wheels project is to help users learn complicated procedures through the use of self-disclosure learning techniques (as opposed to ”help” mechanisms which are separate from actual task execution). Self-disclosure techniques will be used by implementing animation and verbal explanation methods.

      Our project will focus on using these self-disclosure techniques to explain the visualsyntax; a set of visual mappings (rules that define how a data value is represented as acorresponding visual element), and the visual semantics; the insight that each individualvisualization seeks to convey to the user of a system.

      We will build our self-disclosure techniques on learnability literature surveys. From this literature and experimental design we will develop visualizations which create situated learning opportunities embedded within task execution.

      The visual mappings (syntax) and interpretation of visualization (semantics) will reveal themselves to the user during the execution of the task and will slowly disappear as the user progress their skill level. This is analogous to the training wheels being removed from a bicycle, thus the title of the project.

      The end product of the project is to develop a training wheels version of some frequently used visualizations like maps, scatter-plots and hiostograms. By conducting appropriate tests and surveys, we will measure the success of our project and compare the effectiveness of the self-disclosure learning techniques with the traditional help menu based techniques.

Group 4:

Title: Performance Evaluation of Visualization Approaches on High Resolution Displays

Members: Cyril Montabert, Suraj Menon,  Manas Tungare

Link: http://infovis.manastungare.org/

Abstract: Displaying data all the time on a low resolution screen can easily overload the user with (1) fuzziness (2) too much concentration of data.

      Displaying details on demand on smaller-sized displays requires an additional effort on part of the user to dig into the data.       However, high resolution monitors open new perspectives by supporting a details-everytime approach without making the screen look overwhelming and fuzzy.     We intend to evaluate the usability of a details-on-demand visualization technique as compared to a details-visible-everytime approach in the special case of high resolution displays.      We plan to utilize the newly-assembled 3x3 tiled LCD displays that serve as a prototype for the high resolution screen.

Group 5:

Title: Multi-view Visulaization vs. Integrated-view Visualization

Members: Ying Qiao, Bing Fang, Yamin Wu

Abstract: Significant researches have been made on exploring integrated-view vsualization and multi-view visualization, comparisons and analysis have been made about weakness and strength of visualization of different models. However, there are many areas left unexplored. With “the guidelines for using multiple views in information Visualization”[1], in reference to research papers[2], written recently by graduate students from computer science Department of Virginia Polytechnic Institute and State University, we intend to explore further areas to compare and analyze weakness and strength of different presentations of views and interaction among views. By presenting view models different from those research papers, we use the same task scenarios to perform usability experiment as listed in the research paper on geographical map. In addition, we further probe the mental models by generating new tasks that users have in multiple view and integrated view visualizations. Based on comparison of results gained from presenting different models, we conclude that one view model is favorable over the other.

1: Michelle Q. Wang Baldonado, Allison Woondruff, Allan Kuchinsky “Guidelines for Using Multiple Views in Information Visualization”
2: Pak Chung, Xiaoyan Yu, Yogita Bhardwj, Chris North “Empirical Evaluation of Integrated View vs. Multiple Views Visualization on Geographic Maps

 Group 6:

Title: Multiple-step-aggregation Visualization

Members: Hui Yang, Tao Ni

Abstract: It is a common task when analyzing a large amount of data (e.g. Census databases) to create some kind of the overview of the original dataset, which is small enough to be easily manipulated, while remains the key characteristics of the data. The traditional SQL statement provides GROUP BY operation to support tuple aggregation, however, common users with little knowledge in database systems often meet difficulties, i.e. they are not aware when and how to use it efficiently to get the desired result. In addition, standard SQL only supports categorical grouping, while other common grouping types in information exploration, such as range grouping, can't be implemented directly. Finally, it is very easy to get lost after a set of aggregation operations in a huge dataset. Our goal is to design a navigation tool to help users obtain various levels of overviews to narrow their selection, as well as effectively visualizing the aggregation process to increase context awareness. In particular, we consider a major improvement toward Conklin's polyarchy table visualization with augmented aggregation component, and applied it to the census dataset.

Group 7:

Title: A Tool for Visualization of Knowledge Integration, Schema Mapper

Members: Raghavan, Ananth; Rangarajan, Divya

Link: http://filebox.vt.edu/users/ananthr/InfoViz

Abstract: This project proposes to design and develop a tool that will enable visualization of knowledge integration. Specifically this comprises of

1)      Knowledge structure visualization by visualizing mapping of local database schemas to a global schema. This greatly simplifies this process of mapping because it helps the user understand the structure of data involved instantly. The tool will allow the user to create and store mapping rules directly from the visualized schemas rather than from code

2)      Knowledge base visualization by creating a knowledge base which tracks the mapping process by storing the rules of mapping and visualize the same with a view of educating the user. The user will be able to see a complete view of the existing mapping rules in the knowledge base for the global schema and make decisions about creating new rules as and when integrating new schemas

The significant advantage of creation of this knowledge base is knowledge sharing.

 

 Group 8:

Title: Visualization of Clustered traffic on the Network

Members: Brian Badillo, Yoon-Soo Lee

Abstract: