Abstract:Due to the influence of the customers dialect,and the form of customers’addresses that operators of power electrical call center manually record are diverse,it is difficult to accurately find the similar addresses in village level and locate frequent blackouts. In this paper, a fuzzy address matching model is proposed. Based on the character and pinyin forms of address information, similar addresses can be effectively located by the minimum edit distance algorithm and Support Vector Machine(SVM) classifier. The simulation results show that the method can overcome the influence of homophonic words with fast calculation and high reliability. Scenario applications such as locating frequent outage addresses can be supported by the method.