Research on data conflict resolution for web data fusion

摘要

Data fusion is the process of combining multiple records representing the same real-world object into a single, consistent, and clean representation. In order to solve data conflict, an approach for resolving data conflict based on Markov logic networks is proposed. First, a data candidate set which have a higher veracity is obtained by assessing the accuracy of the data source. Second, with the aid of the results from the data candidate set, the attributes of which data conflict can be resolved via adding some more rules, such as the relationship between data sources, the relationship between data sources and facts and the relationship between facts. Experimental results show that the proposed approach can resolve the data conflict effectively, which is more accurate.