False invoice issuance identification method based on parallel association rule mining

Abstract

The present invention discloses a false invoice issuance identification method based on parallel association rule mining. The method comprises the following steps: extracting an invoice record withina required accounting period, and constructing a transaction set; mining a frequent item set separately for a transaction subset of details in different industries to generate an association rule; andbased on the association rules of details in different industries, calculating self-consistency of the enterprise, and identifying the enterprise with the false invoice issuance behavior through a self-consistency calculation result. According to the method disclosed by the present invention, program design is performed based on the Spark parallel framework, by using an Apriori algorithm, association rule mining is performed on the industries to which the upstream and downstream enterprises of the enterprise belong, the abnormal industry combination is analyzed, the method can target the operating characteristics of the details in different industries in a fine-grained manner and has the characteristic of efficient parallelism, and the false invoice issuance behavior can be automatically,quickly and effectively identified.

Description

technical field [0001] The invention relates to the technical field of tax control, in particular to a method for identifying false invoices based on parallel association rule mining. Background technique [0002] Invoices play an extremely important role in our country's social and economic activities. Invoices are an important basis for tax authorities to control tax sources and collect taxes. They are also an important means for the state to supervise economic activities, maintain economic order, and protect national property security. False invoices are objectively manifested as issuing special value-added tax invoices or other invoices for export tax refunds or tax credits for others, for oneself, for oneself, for oneself, or for introduction of others without the purchase or sale of goods or the provision or acceptance of taxable services. Even if there is a purchase or sale of goods or the provision or acceptance of taxable services, special value-added tax invoices ...

AI Technical Summary

Problems solved by technology

Based on the idea of ​​association rule mining in data mining, a parallel Apriori algorithm is proposed to mine association rules for the industries of upstream and downstream enterprises of the target enterprise. By constructing transaction sets, mining frequent item sets, generating association rules, and calculating self-consistency, Then identify false invoices to solve the problems of inaccurate positioning of false invoices in the method of background technology, dependence on sample sets, and limitations of identification types

Images