| The amount of Semantic Web data is growing rapidly today. Individual users, academic institutions and businesses have already published and are continuing to publish their data in Semantic Web standards, such as RDF and OWL. Due to the decentralized nature of the Semantic Web, the same real world entity may be described in various data sources with different ontologies and assigned syntactically distinct identifiers. Furthermore, data published by each individual publisher may not be complete. This situation makes it difficult for end users to consume the available Semantic Web data effectively. In order to facilitate data utilization and consumption in the Semantic Web, without compromising the freedom of people to publish their data, one critical problem is to appropriately interlink such heterogeneous data. This interlinking process is sometimes referred to as Entity Coreference, i.e., finding which identifiers refer to the same real world entity. In the Semantic Web, the owl:sameAs predicate is used to link two equivalent (coreferent) ontology instances. An important question is where these owl:sameAs links come from. Although manual interlinking is possible on small scales, when dealing with large-scale datasets (e.g., millions of ontology instances), automated linking becomes necessary. This dissertation summarizes contributions to several aspects of entity coreference research in the Semantic Web. First of all, by developing the EPWNG algorithm, we advance the performance of the state-of-the-art by 1% to 4%. EPWNG finds coreferent ontology instances from different data sources by comparing every pair of instances and focuses on achieving high precision and recall by appropriately collecting and utilizing instance context information domain-independently. We further propose a sampling and utility function based context pruning technique, which provides a runtime speedup factor of 30 to 75. Furthermore, we develop an on-the-fly candidate selection algorithm, P-EPWNG, that enables the coreference process to run 2 to 18 times faster than the state-of-the-art on up to 1 million instances while only making a small sacrifice in the coreference F1-scores. This is achieved by utilizing the matching histories of the instances to prune instance pairs that are not likely to be coreferent. We also propose Offline, another candidate selection algorithm, that not only provides similar runtime speedup to P-EPWNG but also helps to achieve higher candidate selection and coreference F1-scores due to its more accurate filtering of true negatives. Different from P-EPWNG, Offline pre-selects candidate pairs by only comparing their partial context information that is selected in an unsupervised, automatic and domain-independent manner. In order to be able to handle really heterogeneous datasets, a mechanism for automatically determining predicate comparability is proposed. Combing this property matching approach with EPWNG and Offline, our system outperforms state-of-the-art algorithms on the 2012 Billion Triples Challenge dataset on up to 2 million instances for both coreference F1-score and runtime. An interesting project, where we apply the EPWNG algorithm for assisting cervical cancer screening, is discussed in detail. By applying our algorithm to a combination of different patient clinical test results and biographic information, we achieve higher accuracy compared to its ablations. We end this dissertation with the discussion of promising and challenging future work. |
