Study of Record Matching Algorithm

  • Guannan Wang
  • Published 2015


Every day of our lives, each of us is bombarded with an explosion of data, so that it is nearly impossible to ignore the increasing volume of and potential uses for Big Data. To tackle these difficulties, parallel and distributed computing platforms of Big Data are being considered to handle various aspects of large quantities of data. While considerable progress has been made in improving such tools, studies in many scientific areas remain an elusive challenge; performance of record matching is one of them. Nowadays, a great challenge that many insurance companies have been facing is performing the match of an insurance company’s records to the huge Social Security Death Master File/Database (DMF) in order to verify their policy payment. In this work, a fast, efficient and accurate matching algorithm is developed to match the records from insurance companies to the DMF. The finite sample behavior of the proposed algorithm is evaluated via several simulation studies. Index words: record matching, searching algorithm, shingles encoding,

Cite this paper

@inproceedings{Wang2015StudyOR, title={Study of Record Matching Algorithm}, author={Guannan Wang}, year={2015} }