Bridging cognitive gaps between user and model in interactive dimension reduction
[2];
- Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Blacksburg, VA (United States). Computer Science
- Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Blacksburg, VA (United States). Computer Science; Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Blacksburg, VA (United States). Statistics
Interactive machine learning (ML) systems are difficult to design because of the "Two Black Boxes" problem that exists at the interface between human and machine. Many algorithms that are used in interactive ML systems are black boxes that are presented to users, while the human cognition represents a second black box that can be difficult for the algorithm to interpret. These two black boxes create cognitive gaps between the user and the interactive ML model when a user interacts with the system. In this paper, we identify several cognitive gaps that exist in a previously-developed interactive visual analytics (VA) system, Andromeda. These cognitive gaps that we are addressing in Andromeda are representative of common problems in other VA systems. Our goal with this work is to open both black boxes and bridge these cognitive gaps by making improvements to the original Andromeda system, including designing new visual features to help people better understand how Andromeda processes and interacts with data and improving the underlying algorithm so that the Andromeda system can better understand the intent of the user during the data exploration process. We evaluate our designs through both qualitative and quantitative analysis (i.e., user study and simulation analysis), and the results confirm that the improved Andromeda system outperforms the original version significantly in a series of high-dimensional data understanding tasks.
- Research Organization:
- Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- Sponsoring Organization:
- USDOE
- Grant/Contract Number:
- AC05-76RL01830
- OSTI ID:
- 1785627
- Report Number(s):
- PNNL-SA-157445
- Journal Information:
- Visual Informatics, Vol. 5, Issue 2; ISSN 2468-502X
- Publisher:
- Elsevier - Zhejiang University PressCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Similar Records
Evaluation of Visual Analytics Environments: The Road to the Visual Analytics Science and Technology Challenge Evaluation Methodology
Real-Time Canister Welding Health Monitoring and Prediction System for Spent Fuel Dry Storage