PROPOSED PROGRAMME OF WORK
I. SPECIFIC TASKS
1. The following tasks have been identified:
A. Task 1 - Expertise database construction
Develop a database of related visualization work. This should include contact names and a taxonomy of topics under study. The database will be complete during the first six months, but will be updated continually during the life of the RSG, and made available on-line internationally. [This is considered to be a needed resource for the research outlines in steps 4 and 5]. A subtask in this activity is to develop and circulate a network of experts working in this field and related ones.
B. Task 2 - Test database construction
Assemble a large, highly varied test database of military text, from various sources and derived from several media -including real or simulated automatic speech recognition. [Not only should this be large and varied, it must be capable of simulating the rapid change and input that will characterize dataflow in fielded applications].
C. Task 3 - Prototype Integration
Assemble flexible visual interface software to facilitate search and navigation in a text corpus. (This will be an advanced analysis capability for analysts' workstations, possibly including a Virtual Reality interface). Participating Nations will develop or adapt existing segments of software to perform specific functions. They will share the developed software among themselves, for use and testing as the work progresses.
D. Task 4 - Test clustering/search methods
Develop automated techniques for structuring and visualizing data and supporting decision making. Realize promising techniques in the interface. Test the best strategies in search through the test dataset [and other corpora if available].
E. Task 5 - Develop methods
Develop automated techniques for information extraction and coordination. [Experts among the Workshop participants expect that this will be ongoing research, not complete within two years].
II. DELIVERABLES
Year 1: Two databases. Network of experts
Year 2/3: Review Milestone Report of the RSG, detailing:
- Development of the visualization database, including descriptive metadata
- Development of the test database, including major descriptors
- Progress in seeking or fabricating the elements for the Flexible Visual Interface prototype
- Progress on research Tasks 4 [data structuring and search techniques], and 5 [automated information extraction and- coordination].
Year 4: Final Report of the Group for the work period, detailing:
- Prototype integration
- Results of visual search against text corpora
- Updated metadata summarizing the current contents of the databases, developed under tasks 1 and 2
- Final status of research task 4
- Status of ongoing research task 5
- Prototype flexible visual interface, fielded in selected applications.
III. SCHEDULE
2. The planned schedule is shown on the attached bar chart.
IV. CONTRIBUTION
3. Canada, Denmark, The Netherlands, United Kingdom and United States have expressed an interest in participating in this work. Canada is willing to lead.
V. STATUS
4. The status of this RSG is to be OPEN. The results from this work will be richer as a function of the number of participating nations and their level of effort.