Large model-based product normalization

Abstract

The present application discloses a large model-based product normalization method and system, a storage medium, and a computer device. The method comprises: acquiring product titles and custom product codes corresponding to a plurality of products; clustering the product titles to determine a product category, determining a target product code corresponding to the product category according to the custom product codes of the various products in the product category, and filtering, from among the various products in the product category and according to a product code cluster to which the target product code belongs, products having custom product codes belonging to the product code cluster, wherein products corresponding to respective product codes in the product code cluster are the same product; determining, from among the various products in the product category, a first product to be normalized whose custom product code does not belong to the product code cluster, and performing product normalization for the first product to be normalized by means of a large model according to a product title corresponding to the first product to be normalized and a standard product title in a standard product library, so as to determine a correct product code corresponding to the first product to be normalized. The present application achieves efficient and accurate normalization for products.

Images