INDEX: intention and explanation for fusion of uncertain, noisy, and dynamic spatial data
Key Points
- DAIRNet Theme: Patterns in Noisy and Dynamic Data
- Organisation: RMIT University
- Point of contact: Prof Matt Duckham
The INDEX project is about supporting more effective action from noisy and dynamic data streams, by enabling automatic inference of agent intentions as well as understandable explanations of those inferences.Confidence and clarity in decisions requires more than rigorous techniques for high-and low-level fusion of uncertain data; it also requires the ability to support conclusions about the intentions of tracked agents and to justify and explain the reasoning that led to those conclusions.
The INDEX project will create new techniques and tools for automated intent recognition and explanation from fusion of noisy and dynamic, soft and hard spatiotemporal data streams.
publications
- Duckham, M, Gabela, J, Kealy, A, Khan, M, Legg, J, Moran, B, . . . Vasardani, M 2022, ‘Explainable spatiotemporal reasoning for geospatial intelligence applications’, Transactions in GIS, vol. 26, pp. 2455-2479. DOI: 10.1111/tgis.12939
