Print Email Facebook Twitter Probabilistic record linkage with the Fellegi and Sunter framework: Using probabilistic record linkage to link privacy preserved police and hospital road accident records Title Probabilistic record linkage with the Fellegi and Sunter framework: Using probabilistic record linkage to link privacy preserved police and hospital road accident records Author De Bruin, J. Contributor Jongbloed, G. (mentor) Faculty Electrical Engineering, Mathematics and Computer Science Department Applied mathematics Date 2015-10-28 Abstract Record linkage is the procedure of bringing together information from two or more records that are believed to belong to the same entity. The linking of a pair of records without identifier should be based on attributes both records have in common. The probabilistic framework described by Fellegi and Sunter (1969) can be used for record linkage. This record linkage framework classifies pairs of records as links, non-links or possible links based on a comparison of the attributes found in both records. In this thesis, the framework is studied and methods to estimate the parameters of the framework. The framework is applied to link privacy preserved police and hospital road accident records. The linkage of these data sources is of great interest in the field of road safety research. Subject record linkagedata matchingdeduplicationfellegi and sunterem-algorithmstatisticsprobabilistic record linkagedeterministic record linkageclassificationindexingdatabases To reference this document use: http://resolver.tudelft.nl/uuid:4ebdb50d-c02f-430b-8037-319b45a3e5db Part of collection Student theses Document type master thesis Rights (c) 2015 De Bruin, J. Files PDF Thesis_master_FellegiSunter.pdf 9.19 MB Close viewer /islandora/object/uuid:4ebdb50d-c02f-430b-8037-319b45a3e5db/datastream/OBJ/view