Invoice and goods classifying method based on heuristic rule

Abstract

The invention discloses an invoice and goods classifying method based on a heuristic rule. Non-standardized goods names on invoices are classified into a subcategory in national economy industry categories, and the classification result can directly reflect the operational information of an invoice issuing enterprise. On one hand, whether issued invoices are legal or not can be judged by comparing the invoices with the industry code of the enterprise during register; on the other hand, more visual information and more careful monitoring strength are provided for tax payment risk monitoring of the enterprise. The invoice goods names are classified into customhouse codes first through the case matching of customhouse historical records, the mapping relation from the customhouse codes to national economy industry category codes is artificially established, and finally the national economy industry category codes of the invoices are obtained. Thus, the classification problem that under classification case lacking condition, a large number of non-standardized goods names are classified into the national economy industry categories is solved, and a foundation is laid for subsequent tax analysis and risk monitoring.

Description

technical field [0001] The invention belongs to a method for classifying invoice goods, in particular to a method for classifying invoice goods based on heuristic rules. Background technique [0002] An important data source for tax analysis is the value-added tax invoice of the enterprise. The filling of the goods name mainly includes three items: brand, name and specification model. However, the filling of the goods name in the invoice has not yet been standardized, resulting in uneven quality of the goods data in the invoice. , There are even substandard formats and intentional staggering, which seriously cover up the business information that can be reflected in the invoices, provide a hotbed for tax evasion, and also affect the accuracy of tax analysis. Classifying a large number of invoice goods names has become an urgent problem to be solved in tax analysis and risk monitoring. [0003] At present, there is no classification method from invoice goods to national econ...

AI Technical Summary

Problems solved by technology

Therefore, there are the following problems when the methods described in the above documents are applied to the classification of invoice goods: the classification categories of the classification method of document 1 are self-organized, and it is difficult to control the total number of categories and the coverage level; the classification methods of documents 2 and 3 need to rely on a large number of Classified instance

Images