Weighted Matching of Object Characteristics (Method)


Weighted matching is applied to match two data sets with many common units, on common object characteristics. The method is able to value the strength of possible (candidate) matches by using matching weights. Weighted matching can be formulated as an optimisation problem, in which the optimal (weighted) sum of matches is calculated, under certain constraints, such as that each record can appear in at most one match. The goal of the method is to find solutions to such problems, exact ones or good approximations. The reader is advised to consult the theme module “Micro-Fusion – Object Matching (Record Linkage)” prior to reading the present one. Also the reader should refer to the method module “Micro-Fusion – Unweighted Matching of Object Characteristics”, which can be viewed as a special case of the matching method described in the present paper. It also introduces some concepts that are not re-introduced in the current module.


To read the entire document, please access the pdf file (link under "Related Documents" on the right-hand-side of this page).


Your feedback is appreciated. Please send your remarks, suggestions for improvement, etc. to memobust@cbs.nl.