Abstract

Reference reconciliation is the problem of identifying when different references (i.e., sets of attribute values) in a dataset correspond to the same real-world entity. Most previous literature assumed references to a single class that had a fair number of attributes (e.g., research publications). We consider complex information spaces: our references belong to multiple related classes and each reference may have very few attribute values. A prime example of such a space is Personal Information Management, where the goal is to provide a coherent view of all the information on one’s desktop. Our reconciliation algorithm has three principal features. First, we exploit the associations between references to design new methods for reference comparison. Second, we propagate information between reconciliation decisions to accumulate positive and negative evidences. Third, we gradually enrich references by merging attribute values. Our experiments show that (1) we considerably improve precision and recall over standard methods on a diverse set of personal information datasets, and (2) there are advantages to using our algorithm even on a standard citation dataset benchmark. 1.

Citations

311	A theory for record linkage - Fellegi, Sunter - 1969
254	Merge/Purge Problem for Large Databases - Hernandez, Stolfo - 1995
200	Efficient clustering of high-dimensional data sets with application to reference matching - McCallum, Nigam, et al. - 2000
191	Stuff I’ve seen: a system for personal information retrieval and re-use - Dumais, Cutrell, et al.
172	WE: The state of record linkage and current research problems - Winkler - 1999
161	Interactive deduplication using active learning - Sarawagi, Bhamidipaty - 2002
141	Automating the construction of internet portals with machine learning - AK, Nigam, et al.
130	Identity uncertainty and citation matching - Pasula, Marthi, et al. - 2002
107	Haystack: A Platform for Authoring End User Semantic Web Applications - Quan, Huynh, et al. - 2003
106	Automatic linkage of vital records - Newcombe, Kennedy, et al. - 1959
92	Learning domain-independent string transformation weights for high accuracy object identification - Tejada, Knoblock, et al. - 2002
86	Declarative data cleaning: Language, model, and algorithms - Galhardas, Florescu, et al. - 2001
59	Exploiting relationships for domain-independent data cleaning - KALASHNIKOV, MEHROTRA, et al. - 2005
52	Toward conditional models of identity uncertainty with application to proper noun coreference - McCallum, Wellner - 2003
47	P.: Multi-relational record linkage - Parag - 2004
38	Using the EM algorithm for weight computation in the Fellegi-Sunter Model of record linkage - Winkler - 1988
37	IntelliClean: a knowledge-based intelligent data cleaner - Lee, Ling, et al. - 2000
27	Record linkage: Current practice and future directions - Gu, Baxter, et al. - 2003
11	Semex: Toward on-the-fly personal information integration - Dong, Halevy, et al. - 2004
10	Methods for linking and mining massive heterogeneous databases - Pinheiro, Sun - 1998
6	Exploiting secondary sources for unsupervised record linkage - Michalowski, Thakkar, et al. - 2004