Integration of Probabilistic Uncertain Information

TL;DR

This paper addresses the challenge of integrating probabilistic uncertain data from multiple sources, proposing methods to accurately determine the resulting probability distribution under specific conditions.

Contribution

It introduces a subclass of probabilistic relations allowing exact probability distribution computation after data integration, improving over previous range-based methods.

Findings

Exact probability distributions can be computed for certain subclasses of probabilistic relations.

The proposed approach enhances the accuracy of data integration in probabilistic databases.

Efficient algorithms are developed for the integration process under the new model.

Abstract

We study the problem of data integration from sources that contain probabilistic uncertain information. Data is modeled by possible-worlds with probability distribution, compactly represented in the probabilistic relation model. Integration is achieved efficiently using the extended probabilistic relation model. We study the problem of determining the probability distribution of the integration result. It has been shown that, in general, only probability ranges can be determined for the result of integration. In this paper we concentrate on a subclass of extended probabilistic relations, those that are obtainable through integration. We show that under intuitive and reasonable assumptions we can determine the exact probability distribution of the result of integration.