Abstract | ||
---|---|---|
This paper studies the problem of integrating probabilistic uncertain information. Certain constraints are imposed by the semantics of integration, but there is no guarantee that they are satisfied in practical situations. We present a Bayesian-based approach to revise the probability distribution of the information in the sources in a systematic way to remedy this difficulty. The revision step is similar in spirit to tasks like data cleaning and record linkage and should be carried out before integration can be achieved for probabilistic uncertain data. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1007/978-3-319-19548-3_7 | DATABASES THEORY AND APPLICATIONS |
Keywords | Field | DocType |
Information integration, Uncertain data, Probabilistic data, Belief revision | Record linkage,Information integration,Data mining,Computer science,Uncertain data,Probability distribution,Probabilistic logic,Belief revision,Semantics,Bayesian probability | Conference |
Volume | ISSN | Citations |
9093 | 0302-9743 | 1 |
PageRank | References | Authors |
0.36 | 23 | 1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Fereidoon Sadri | 1 | 846 | 283.70 |