Format mediation for matching patient records in a three-layer information architecture

Many problems are encountered when information is combined from multiple databases. The major concern is the mismatch encountered in representing and structuring the information. Matching these data together is problematic because even small errors, such as data entry errors, different text format, and missing data can prevent the exact-match. Using this information in an intelligent and active manner requires a software module with a mediation role between the user level applications and the databases. The main features of the mediation level are to simplify, abstract, reduce, merge, and explain data. This paper describes a three mediation architecture that can be applied in the healthcare institution. Our focus will be on the mediation component, also known as the mediator. One of the most important features of the mediator is the conflict resolution for which we propose the algorithm Jaro-Winkler, a powerful method for comparing the similarities of short strings.