Reference: Zeng, H.; Fikes, R. Explaining Data Incompleteness in Knowledge Aggregation. Technical Report, Knowledge Systems, AI Laboratory, Stanford University. 2005.
Abstract: Knowledge aggregation is the problem of taking information from multiple heterogeneous sources and aggregating it into a unified knowledge base. One of the main challenges in that work has been dealing with data incompleteness because data sources seldom contain complete answers to a user's query. Current approaches leverage users' preferences over data sources when trying to aggregate incomplete data. Nevertheless, these approaches are not adequate to satisfy users' needs to trust aggregated data before they can use them with confidence in the presence of incomplete information. We believe such trust may be earned by providing users with the explanations for incomplete data. In this paper, we construct a decision tree-based classifier to acquire context knowledge about data sources and build an aggregation system capable of explaining incomplete data with learned context knowledge. Further, our approach provides a new method to characterize sources that may help users better understand the discrepancies between sources.
Full paper available as pdf.