Invoice over-limit application approval method and device, equipment and medium

Abstract

The embodiment of the invention relates to an invoice over-limit application approval method and device, equipment and a medium. The method comprises: obtaining entity information and tax-related data of a to-be-approved entity, wherein the to-be-approved entity is an entity initiating an invoice over-limit application; constructing a first basic feature vector based on the entity information and the tax-related data; generating a first derivative feature vector based on the first basic feature vector, wherein the first derivative feature vector is used for reflecting the distinguishing degree of the to-be-approved entity belonging to an approval passing type or an approval rejection type; and inputting the first basic feature vector and the first derivative feature vector into a pre-trained target application approval model, and determining an approval result of the invoice over-limit application initiated for the entity to be approved based on a model output result. According to the technical scheme, the generalization performance and the approval accuracy of the approval scheme are improved.

Description

technical field [0001] The present invention relates to the field of financial technology, in particular to a method, device, equipment and medium for application and approval of an invoice exceeding the limit. Background technique [0002] Invoice over-limit application refers to the taxpaying entity (such as an enterprise) being affected by various factors, in the actual production and operation process, the actual amount of invoices used exceeds the number of invoices approved by the type of invoice, and then the excess invoice is initiated by the competent tax authority. Application. Approval of application for exceeding the limit of invoices refers to the approval process of the competent tax authority for the application of exceeding the limit of invoices. In order to improve the approval efficiency of the competent tax authorities, computer technology is often used to process relevant data to achieve intelligent / automatic approval. [0003] At present, the main impl...

AI Technical Summary

Problems solved by technology

[0004] However, in the above-mentioned approval scheme based on the expert evaluation index system, the selection of evaluation indicators, the setting of index weights and risk qualitative thresholds all depend on the field experience of tax experts, which makes the scheme have poor generalization performance and poor scalability. bad question

Moreover, the determination strategies for the above three items are all artificially set, which is difficult to adapt to the changes and updates of tax policies, and also difficult to adapt to the changes and updates of corporate behavior data caused by the development of enterprises. Both the adaptability and the adaptability of the enterprise development are poor, which ultimately leads to the low accuracy of the intelligent approval of the scheme

Images