Semantic Perception: Converting Sensory Observations to Abstractions
Year: 2012
Type: article
Abstract: An abstraction is a representation of an environment derived from sensor observation data. Generating an abstraction requires inferring explanations from an incomplete set of observations (often from the Web) and updating these explanations on the basis of new information. This process must be fast and efficient. The authors' approach overcomes these challenges to systematically derive abstractions from observations. The approach models perception through the integration of an abductive logic framework called Parsimonious Covering Theory with Semantic Web technologies. The authors demonstrate this approach's utility and scalability through use cases in the healthcare and weather domains.
Source: IEEE Internet Computing
DOI: 10.1109/mic.2012.20
Institution: Wright State University
Language: English
Topic: Semantic Web and Ontologies
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2141273062