Product Quality Risk Judgment Method Based on Label Recognition

Abstract

The invention provides a commodity quality risk judgment method based on label identification, which comprises the following steps: reading the label information of a commodity, converting the name ofthe commodity into a numerical code set, and carrying out 0-1 vectorized representation on a commodity formula; calculating the comprehensive distance between every two commodities, and adopting K-Medode algorithm to cluster to obtain a commodity class; A violation additive set is determined in the commodity according to the occurrence frequency of ingredients, average mutual information betweenthe ingredients is calculated for each commodity, a certain number of ingredients with the minimum average mutual information are selected, and comparison judgment is conducted on the ingredients andthe violation additive set. The method comprises the steps of commodity clustering. According to the method, the commodity is classified, violation additives are identified, commodity information doesnot need to be registered in advance or a large number of rules do not need to be coded, the violation additives in the commodity formula can be automatically identified and memorized, the learning ability is achieved, the method can be compatible with multiple languages, and rapid identification of the commodity violation additives and automatic screening of commodity quality risks are achieved.

Description

technical field [0001] The invention relates to the technical field of commodity quality analysis, in particular to a method for judging commodity quality risk based on label identification. Background technique [0002] Commodity quality is related to people's life and property safety and is an important area of ​​government supervision. At present, the rapid development of imported cross-border e-commerce has made a large number of goods produced and sold abroad enter my country quickly. This kind of cross-border e-commerce import trade is characterized by small batches and multiple batches, which has caused great pressure on the customs and other regulatory departments. Since the text on foreign product labels is in foreign languages, it is difficult to determine the type of product according to domestic standards, so it is very difficult to determine the risk of the product. [0003] The main basis for judging commodity quality risks is the national standards of the Pe...

AI Technical Summary

Problems solved by technology

In China, there are currently units or departments using the "Smart Import and Export Industrial Product Risk Management Information Platform". Although it uses data analysis methods to manage commodity quality risks, its evaluation model is relatively fixed and cannot be expanded and automatically learned

Although the “Risk Assessment Grading Rules in Technology” in the prior art takes into account the inherent risks of the product and the special risks caused by the place of origin, users, etc., the establishment of the rules relies on manual work and the degree of automation is not high

Images