Uncertain data integration with probabilities

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Gayatri Tallur (Creator)
Institution
The University of North Carolina at Greensboro (UNCG )
Web Site: http://library.uncg.edu/
Advisor
Fereidoon Sadri

Abstract: Real world applications that deal with information extraction, such as business intelligence software or sensor data management, must often process data provided with varying degrees of uncertainty. Uncertainty can result from multiple or inconsistent sources, as well as approximate schema mappings. Modeling, managing and integrating uncertain data from multiple sources has been an active area of research in recent years. In particular, data integration systems free the user from the tedious tasks of finding relevant data sources, interacting with each source in isolation using its corresponding interface and combining data from multiple sources by providing a uniform query interface to gain access to the integrated information. Previous work has integrated uncertain data using representation models such as the possible worlds and probabilistic relations. We extend this work by determining the probabilities of possible worlds of an extended probabilistic relation. We also present an algorithm to determine when a given extended probabilistic relation can be obtained by the integration of two probabilistic relations and give the decomposed pairs of probabilistic relations.

Additional Information

Publication
Thesis
Language: English
Date: 2013
Keywords
Database, Integration, Possible world, Probabilistic relation, Probability, Uncertain
Subjects
Probabilistic databases
Probabilities

Email this document to