Address recognition

as one of the important scenarios of natural language processing in big data applications

is an important and highly practical technical means. At present

power enterprises are actively promoting the evolution of big data applications

and the demand of using big data technology to empower the development of power enterprises is also increasing.The address information in many data assets of power enterprises

as the core field of connecting equipment-customer-enterprise

is extremely valuable for analysis and mining. The common address information application scenarios of power enterprises are first analyzed

and aiming at the key points

the extracting and matching methods are formulated for the address recognition requirements of the core application scenes. And based on the data samples

the accuracy and operating speed of the address recognition methods are studied and analyzed

and the practicability of the algorithms is compared. Finally

the algorithms are summarized and the future algorithm improvement methods are prospected

which gives a reference for the practical application to the related businesses of power enterprises.