The IRMA Community
Newsletters
Research IRM
Click a keyword to search titles using our InfoSci-OnDemand powered search:
|
Evidential Characterization of Uncertainty in Location-Based Prediction
Abstract
This paper provides a novel approach to characterization of uncertainty measures in classification and prediction of complex spatial objects in data mining. The paper shows the semantic limit of uncertainty measure in classical probabilistic approaches and presents a formal approach to characterize uncertainty parameters from Rough set and Dempster-Shafer’s evidence theory in spatial domain. We have developed a rough set and Dempster-Shafer’s evidence theory based formalism to objectively represent uncertainty inherent in the process of service discovery, characterization, and classification. Rough set theory is ideally suited for dealing with limited resolution, vague and incomplete information, while Depster-Shafer’s evidence theory provides a consistent approach to model an expert’s belief and ignorance in the classification decision process. Integrating these two approaches provide a mathematically consistent and objective means to measure belief, plausibility, ignorance and other useful measures in spatial classification. Moreover, it provides predictive measures of uncertainty and thereby allows including the context of spatial neighborhood effects.
|
|