A Novel Method for Validating Addresses Using String Distance Metrics

Address validation is vital since it confirms the quality and geographical precision of addresses used by organizations that rely on location-dependent and delivery-based services. Suppose addresses need to be thoroughly checked in advance. In that case, there may be difficulties with them, such as missing components or geographical deficiencies, which may lead to severe problems with logistics. When doing address validation, discovering missing or incorrect address components is a beneficial aspect in minimizing the likelihood of service problems while saving time and money for organizations. When it comes to addressing validation, using statistical metrics like correlation coefficients and measures of central tendency has been discovered to have a significant amount of untapped potential. In order to obtain a normalized score that is based on statistical similarities, the approach that is suggested in this study makes use of a mixture of several string-matching metrics. This score may then be used to exclude authenticated addresses based on the needed minimum level of similarity, which can be calculated. Experiments have been carried out on a healthcare dataset taken from the actual world to show the efficacy of the suggested method in terms of accuracy and precision.