Accessory name recognition method and device, computer equipment and storage medium

Abstract

The invention relates to the field of artificial intelligence, and by performing word segmentation and adding a noun of locality according to an accessory hierarchical recognition model correspondingto an accessory name, the accuracy and efficiency of accessory name recognition can be improved. The invention particularly relates to an accessory name recognition method and device, computer equipment and a storage medium. The accessory name recognition method comprises the steps: acquiring a to-be-recognized accessory name, and determining the type of a vehicle corresponding to the accessory name; determining an accessory recognition library corresponding to the accessory name according to the type of the vehicle corresponding to the accessory name; if the recognition name corresponding tothe accessory name does not exist in the accessory recognition library, determining an accessory hierarchical recognition model corresponding to the accessory name according to the type of the vehiclecorresponding to the accessory name; and according to the accessory hierarchical recognition model, performing word segmentation on the accessory name and adding a preset noun of locality to determine a recognition name corresponding to the accessory name. In addition, the invention also relates to a blockchain technology, and the accessory recognition library and the accessory hierarchical recognition model can be stored in the blockchain.

Description

technical field [0001] The present application relates to the field of artificial intelligence, in particular to a method, device, computer equipment and storage medium for recognizing an accessory name. Background technique [0002] In the automotive aftermarket business, an urgent problem to be solved is the standardization of parts names; for the same parts, different OEMs, repair shops, and 4S shops have different names in different regions, causing various problems. Standardized parts names can not only improve efficiency , and can improve the anti-fraud and anti-leakage identification level of insurance companies, and provide more intelligent and precise fraud prevention and control capabilities. [0003] The inventor found in the research that the identification methods of parts names currently on the market mainly include expert matching systems, fuzzy query systems and the like. Among them, the expert matching system includes human-computer interaction interface, k...

AI Technical Summary

Problems solved by technology

Among them, the expert matching system includes human-computer interaction interface, knowledge base, comprehensive database, reasoning machine, interpreter, knowledge acquisition and other structures; the expert matching system makes the computer recognize the parts name according to the knowledge, experience and skills of human experts. It takes a long time to recognize the name of the accessory, and the recognition efficiency is low

The fuzzy query system is similar to a search engine. The user enters the keyword of the accessory name for fuzzy search, and the matching result is inaccurate or most of the redundant information is searched out, which cannot improve the accuracy and efficiency of accessory name recognition.

Images