Improving approximate matching capabilities for Meta Map Transfer applications

Using approximate matching technique for different application areas ranging from bioinformatics to wireless world is a challenging task. With the developments of new approximate matching techniques, the efficiency and speed are the main factors in deciding among different options available for each application area. Each application area has certain special features that can be used by approximate matching technique best suited for that area. In addition to direct application to information retrieval, the approximate matching techniques can also be used for complex computational procedures involving similarity search.In this paper, we present an approach for extension of Java class gov.nih.nlm.mmtx.ApproximateMatching from java.lang.Object using an approximate matching technique [1] which is based on reversing the conventional scheme of error-correction coding. This technique developed for multi-attribute objects can be applied to Meta Map Transfer application that assumes that the input is a term. As an extension of the Approximate Matching class we introduce the Store and Retrieve extensions which can be used to store and retrieve terms based on approximate matching technique. Mapping options for the Meta Map Transfer application are also discussed using this approximate matching technique.