Fuzzy control structures in multiple parameter systems: anapplication in handwritten address interpretation systems

摘要

Most practical software systems are a conglomerate of manymodules, where each module has its own parameters that control theaccuracy of the module. While each individual module can be optimized bytuning the relevant parameters, it is a non-trivial task to optimize theentire system. When the number of modules and the parameters are few,manual choosing of all parameters is possible by a “trial anderror” mechanism. However, when the modules are many, othermethods have to be adopted. We use fuzzy set technology for the purposein the particular application of a Handwritten Address InterpretationSystem. We use a fuzzy measure based on the topology of the“blind” set of handwritten addresses and a predeterminedneighborhood to construct the fuzzy membership function. This allows anonlinear partitioning of the results to maximize the correct rate. On atest set of 10000 images, the fuzzy methodology accomplishes 5% higheraccuracy