CN112184022A - White spirit consumption tax risk assessment method - Google Patents

White spirit consumption tax risk assessment method Download PDF

Info

Publication number
CN112184022A
CN112184022A CN202011046196.XA CN202011046196A CN112184022A CN 112184022 A CN112184022 A CN 112184022A CN 202011046196 A CN202011046196 A CN 202011046196A CN 112184022 A CN112184022 A CN 112184022A
Authority
CN
China
Prior art keywords
white spirit
enterprise
price
weighted average
taking
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202011046196.XA
Other languages
Chinese (zh)
Inventor
迟敬泽
吴敬
金语泽
朱淞鹤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Digital China Information Systems Co ltd
Original Assignee
Digital China Information Systems Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Digital China Information Systems Co ltd filed Critical Digital China Information Systems Co ltd
Priority to CN202011046196.XA priority Critical patent/CN112184022A/en
Publication of CN112184022A publication Critical patent/CN112184022A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0202—Market predictions or forecasting for commercial activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0206—Price or cost determination based on market factors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/12—Accounting
    • G06Q40/123—Tax preparation or submission

Abstract

The invention provides a risk assessment method for white spirit consumption tax, which comprises the following steps: constructing a liquor dictionary based on three key elements of liquor by utilizing liquor invoice data, and realizing liquor secondary classification and liquor class matching functions through the liquor dictionary; clustering the three elements of the white spirit contained in the invoice, extracting white spirit classes and constructing a white spirit class library; establishing a relationship network by using the tax data; carrying out statistical analysis based on a relationship network, and calculating the delivery price and the sale price of each kind of white spirit according to a calculation rule; comparing the calculated delivery price and sale price of each kind of white spirit with the consumption tax declaration data of the kind of white spirit, evaluating the consumption tax risk of the white spirit, and generating a risk evaluation result report. The method does not depend on manual experience, and automatically calculates the price of the white spirit by analyzing and mining the data related to the tax, and compares the price with the white spirit declaration data, thereby efficiently and accurately evaluating the white spirit consumption tax risk.

Description

White spirit consumption tax risk assessment method
Technical Field
The invention relates to the technical field of data analysis and mining, in particular to a white spirit consumption tax risk assessment method.
Background
The tax bureau receives massive data every year, part of the data is generated when the enterprise generates real tax-corresponding behavior, and the other part of the data is generated when the enterprise steals national tax for self benefit and provides data which is not consistent with actual operation business conditions. Aiming at the white spirit industry, the taxes collected by the tax bureau are mainly consumption taxes. The tax administration sets up white spirit sold by multi-stage sales units for white spirit production enterprises, and the national tax authority shall decide the lowest tax-counting price of the production enterprise for consumption tax according to the final first-stage sales unit. The specific requirements are as follows: 1. the liquor sold to the sales units by the liquor production enterprises, the tax-counting price of the production enterprises is higher than the external sales price of the sales units by more than 70 percent (including 70 percent), and the tax authority temporarily does not check the lowest tax-counting price of the tax consumption. 2. The liquor production enterprises sell the liquor to the sales units, the tax-counting price of the production enterprises is lower than the external sales price of the sales units by less than 70%, and the consumption tax is collected according to the lowest tax-counting price.
At present, the tax bureau mainly evaluates the risk of the white spirit consumption tax by means of manual screening so as to find taxpayers who pay less white spirit consumption tax. However, the manual screening mode depends on the experience of staff in the tax bureau, and in the face of massive and complicated white spirit production enterprises and relevant tax data, the manual screening efficiency is very low, so that the white spirit consumption tax risk cannot be comprehensively and accurately evaluated.
Therefore, aiming at the problems of the tax bureau in the inspection of the white spirit consumption tax, the invention provides the white spirit consumption tax risk assessment method, which constructs a white spirit dictionary and a plurality of relationship networks for statistical analysis and calculation through a data analysis and mining technology, and automatically discovers taxpayers with white spirit consumption tax risks, thereby assisting the tax authority in leaking stoppage and increasing income and reducing the manual workload of tax staff.
Disclosure of Invention
The invention provides a white spirit consumption tax risk assessment method, which aims to solve the problems that in the prior art, the manual screening mode depends on the experience of tax office workers, and the manual screening efficiency is very low in the case of massive and complicated white spirit production enterprises and relevant tax data, so that the white spirit consumption tax risk cannot be assessed comprehensively and accurately.
The invention provides a risk assessment method for white spirit consumption tax, which comprises the following steps:
s1, decomposing white spirit elements in the white spirit goods name through white spirit invoice data, and constructing a white spirit dictionary through the white spirit elements;
s2, performing TF/IDF feature extraction on white spirit elements through a white spirit dictionary, clustering the white spirit elements of all white spirits through a Kmeans clustering algorithm, extracting all white spirit categories and constructing a white spirit category library;
s3, constructing a white spirit tax-related relation network by using tax data;
s4, carrying out statistical analysis on the white spirit tax relationship network, and calculating the delivery price and the sale price of each kind of white spirit;
s5, comparing the calculated delivery price and sale price of each kind of white spirit with the consumption tax declaration data of the kind of white spirit, if the calculated sale price is multiplied by 0.7, the calculated delivery price is larger than the calculated delivery price, and if no product white spirit declaration data exists, the product white spirit is not declared, if the declared delivery price is larger than the declaration lowest tax price or the calculated sale price is multiplied by 0.7, the calculated delivery price is multiplied by 0.6, and the lowest tax price is declared, the product white spirit is false; otherwise, the report is correctly declared.
Firstly, data cleaning is carried out on goods names of liquor invoices, the cleaned data are subjected to word segmentation, three types of parts of speech are newly added in a word segmentation dictionary to respectively represent the names, specifications and degrees of liquor, and corresponding words are added into the dictionary to construct a liquor dictionary. Based on the constructed liquor dictionary, liquor can be extracted from the goods names of any liquor invoices by three key elements. And (4) realizing liquor secondary classification by judging whether the names of the goods contain liquor names or not, and judging whether any goods name belongs to liquor. The liquor types are matched through the comparison of three key factors of different liquors, and whether the names of two liquor goods belong to the liquor of the same type is judged.
According to the method for evaluating the risk of the white spirit consumption tax, which is disclosed by the invention, as an optimal mode, white spirit elements comprise name information, specification information and degree information.
According to the method for evaluating the risk of the white spirit consumption tax, as an optimal mode, a white spirit tax-related relation network comprises a white spirit production enterprise sales network, a white spirit production enterprise consanguineous network and a white spirit retailer sales network;
the white spirit production enterprise sales network takes white spirit sold by white spirit production enterprises as a starting point, and takes the enterprises as nodes, the purchase-sale relationship as directed edges and the invoice amount and quantity as edge weights according to the types of the white spirit;
the method comprises the steps that a white spirit production enterprise blood margin network establishes a non-directional edge with a white spirit production enterprise as an associated blood margin enterprise by taking the white spirit production enterprise as a central node and an enterprise having a common investor relationship with the white spirit production enterprise or an enterprise belonging to the same branch office of the white spirit production enterprise as the associated blood margin enterprise through enterprise common investor and branch office information aiming at the white spirit production enterprise;
the white spirit retailer sales network takes the white spirit purchase of large-scale retailer enterprises as a starting point, the white spirit retailer sales network takes the enterprises as nodes, the purchase-sale relationship is taken as a directed edge, and the amount and quantity of invoices of the white spirit are taken as edge weights.
White spirit manufacturing enterprise sales network: with white spirit sold by white spirit production enterprises as a starting point, aiming at a certain type of white spirit of a certain white spirit production enterprise, whether the goods name of a white spirit invoice belongs to the same type is judged through white spirit type matching, and therefore the purchase-sale relationship of the white spirit of the type based on the enterprise invoice is found.
The blood relationship network of the white spirit production enterprise: aiming at the white spirit production enterprises, the white spirit production enterprises are used as central nodes, enterprises which have the same investors as the white spirit production enterprises in the information of the registered investors or enterprises which have main institutions with the white spirit production enterprises in the information of the registered branches are used as consanguineous enterprises to establish undirected edges with the white spirit production enterprises.
The liquor retailer sales network takes liquor purchased by a large-scale retailer enterprise as a starting point, judges whether the goods name of a liquor invoice belongs to the same type through liquor type matching aiming at a certain type of liquor, thereby finding out the purchase-sale relationship of the liquor based on the enterprise invoice and backtracking upwards from the large-scale retailer as the starting point.
The invention relates to a liquor consumption tax risk assessment method, which is an optimal mode, and the method for calculating the delivery price of the liquor in the step S4 specifically comprises the following steps:
Figure BDA0002708065390000031
wherein P is the factory price of a certain kind of white spirit, N is the total quantity of the kind of white spirit sold by N white spirit production enterprises as sellers, N iskThe amount of white spirit in the kth transaction, UkThe unit price of the white spirit in the kth transaction.
The invention relates to a liquor consumption tax risk assessment method, which is an optimal mode, and the liquor selling price calculation method in the step S4 specifically comprises the following steps:
s421, judging whether the enterprise is the final layer of the analysis starting point enterprise in the white spirit production enterprise sales network, if so, taking the enterprise as a weighted average A1 of a first-layer price of a purchasing party; if not, then no value is assigned to weighted average A1;
s422, judging whether the enterprise is a marketing enterprise of which the analysis starting point enterprise is the last layer in the liquor production enterprise sales network, if so, taking the enterprise as a weighted average A2 of the first-layer price of the marketing enterprise; if not, then no value is assigned to weighted average A2;
s423, judging whether the number of the enterprises sold by the enterprise at a certain node in the white spirit production enterprise sales network is more than N1If yes, taking the enterprise as a weighted average A3 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then no value is assigned to weighted average A3;
s424, judgmentWhether the disconnected enterprise sells more than N enterprises for a certain node enterprise in the white spirit production enterprise sales network2Home and price difference greater than P1If yes, taking the enterprise as a weighted average A4 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then no value is assigned to weighted average A4;
s425, judging whether the selling price of the enterprise at a certain node in the selling network of the white spirit production enterprise is increased more than P2If yes, taking the enterprise as a weighted average A5 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then no value is assigned to weighted average A5;
s426, judging whether the enterprise is the final layer of the analysis starting point enterprise in the blood relationship network of the white spirit production enterprise, if so, taking the enterprise as a weighted average B1 of one-layer prices of the purchasing parties; if not, no value is assigned to weighted average B1;
s427, judging whether the enterprise is a marketing enterprise with an analysis starting point enterprise as the last layer in the blood margin network of the white spirit production enterprise, if so, taking the enterprise as a weighted average B2 of the first-layer price of the marketing enterprise; if not, no value is assigned to weighted average B2;
s428, whether more than N enterprises are sold by a certain node enterprise in the blood supply network of the white spirit production enterprise or not is judged1If yes, taking the enterprise as a weighted average of the one-level price of the seller B3, and taking the enterprise result closest to the starting point when there are a plurality of such enterprises; if not, no value is assigned to weighted average B3;
s429, judging whether the number of the enterprises sold by a certain node enterprise in the blood relationship network of the white spirit production enterprise is more than N2Home and price difference greater than P1If so, taking the enterprise as a weighted average of the one-tier price of the seller B4, and taking the enterprise result closest to the starting point when there are a plurality of such enterprises; if not, no value is assigned to weighted average B4;
s4210, judging whether the price increase of a certain node enterprise in the blood relationship network of the white spirit production enterprise is more than P2If yes, take the enterprise as the sellerA weighted average of the prices at one level B5, taking the business result closest to the origin when there are multiple such businesses; if not, no value is assigned to weighted average B5;
s4211, acquiring a weighted average C of prices sold to retailers by all consanguineous relation enterprises of the white spirit production enterprises in a white spirit retailer sales network;
s4112, when A3 and A4 exist at the same time, retaining the A3 value; when both B3, B4 are present, the B3 value is retained;
s4113, comparing the weighted average results A1, A2, A3, A4 and A5 processed in the step S4112, taking the highest value as a first selling price, and taking the factory price of the selling network of the white spirit production enterprise as the first selling price if all A1, A2, A3, A4 and A5 are not assigned with values; taking the highest value of the weighted average results B1, B2, B3, B4 and B5 as a second selling price, and taking the factory price of the blood relationship network of the white spirit manufacturing enterprise as the second selling price if no value is assigned to B1, B2, B3, B4 and B5; and C is taken as a third selling price, the first selling price, the second selling price and the third selling price are compared, and the highest value is taken as the final selling price.
And comparing the calculated factory price and sale price of the white spirit with the factory price and the lowest tax-counting price of the white spirit declared by the white spirit production enterprise, dividing the white spirit declared by the enterprise into a correct declaration type, a missing declaration type and an erroneous declaration type, wherein the missing declaration type and the erroneous declaration type have white spirit consumption tax risks, and generating a corresponding risk assessment report according to a comparison result.
According to the liquor consumption tax risk assessment method, as an optimal mode, a liquor dictionary is used for extracting liquor elements from the goods names of liquor invoices, and judging whether the goods names belong to liquor or not according to the fact that the goods names contain liquor names; the liquor types are matched through the comparison of different liquor factors, and whether the names of two liquor goods belong to the liquor of the same type is judged.
The invention relates to a liquor consumption tax risk assessment method, which is an optimal mode, wherein N is1、N2Are integers which are more than or equal to 2 and less than or equal to 10; p1、P2Are all greater than or equal to 0.1 and less than or equal to 1.
Wherein N is1N2As a business parameter, P1P2As a price parameter
The invention has the following beneficial effects:
(1) the method for evaluating the white spirit consumption tax risk does not depend on manual experience, and automatically calculates the delivery price and the sale price of the white spirit by analyzing and mining tax-related data and comparing the delivery price and the sale price with white spirit declaration data, thereby efficiently and accurately evaluating the white spirit consumption tax risk;
(2) the goods names of the white spirit invoices are analyzed based on the three key elements of the white spirit, so that the effects of white spirit secondary classification and white spirit class matching are effectively improved;
(3) building three relation networks of a white spirit production enterprise sales network, a white spirit production enterprise consanguineous network and a white spirit retailer sales network, and fully utilizing the existing related tax data to comprehensively and comprehensively analyze the price of white spirit;
(4) and analyzing statistical rules and formulating a white spirit price calculation rule based on a relational network, so that the accuracy of the calculation of the white spirit price is further improved.
Drawings
Fig. 1 is a flow chart of a liquor consumption tax risk assessment method.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
Example 1
As shown in fig. 1, a method for assessing risk of white spirit consumption tax comprises the following steps:
s1, decomposing white spirit elements in the white spirit goods name through white spirit invoice data, and constructing a white spirit dictionary through the white spirit elements;
s2, performing TF/IDF feature extraction on white spirit elements through a white spirit dictionary, clustering the white spirit elements of all white spirits through a Kmeans clustering algorithm, extracting all white spirit categories and constructing a white spirit category library;
s3, constructing a white spirit tax-related relation network by using tax data;
s4, carrying out statistical analysis on the white spirit tax relationship network, and calculating the delivery price and the sale price of each kind of white spirit;
s5, comparing the calculated delivery price and sale price of each kind of white spirit with the consumption tax declaration data of the kind of white spirit, if the calculated sale price is multiplied by 0.7, the calculated delivery price is larger than the calculated delivery price, and if no product white spirit declaration data exists, the product white spirit is not declared, if the declared delivery price is larger than the declaration lowest tax price or the calculated sale price is multiplied by 0.7, the calculated delivery price is multiplied by 0.6, and the lowest tax price is declared, the product white spirit is false; otherwise, the report is correctly declared.
The specific implementation content of step S1 includes:
the goods names of the screened white spirit invoices are subjected to data cleaning, including removing weight, deleting stop words and deleting irrelevant symbols, and the goods names of the cleaned white spirit invoices can be decomposed into three elements: the name, specification and degree of the product, if the name of the product is 56 degrees Yongfeng green bottle Erguotou 248ml, can be decomposed into: the product is named as Yongfeng green bottle Erguotou, the specification is 248ml, and the degree is 56 degrees;
dividing the goods name of the cleaned white spirit invoice by using a word dividing tool, adding three parts of speech in a word dividing dictionary to respectively represent the name, specification and degree of the white spirit, and adding words belonging to the three parts of speech into the dictionary;
based on the constructed liquor dictionary, the goods names of any liquor invoice can be participled, and three-factor extraction of liquor is realized;
white spirit secondary classification can be realized by judging whether the names of the goods contain white spirit names, namely whether any goods name belongs to white spirit;
through the contrast of the white spirit three key elements of the goods names of different white spirit invoices, the matching of white spirit grades is realized, and whether the two white spirit goods names belong to the white spirit of the same grade or not is judged.
The specific factory price and selling price calculation method in step S4 includes: the factory price calculation method specifically comprises the following steps:
Figure BDA0002708065390000071
wherein P is the factory price of a certain kind of white spirit, N is the total quantity of the kind of white spirit sold by N white spirit production enterprises as sellers, N iskThe amount of white spirit in the kth transaction, UkThe unit price of the white spirit in the kth transaction.
The method for calculating the sale price specifically comprises the following steps:
s421, judging whether the enterprise is the final layer of the analysis starting point enterprise in the white spirit production enterprise sales network, if so, taking the enterprise as a weighted average A1 of a first-layer price of a purchasing party; if not, then no value is assigned to weighted average A1;
s422, judging whether the enterprise is a marketing enterprise of which the analysis starting point enterprise is the last layer in the liquor production enterprise sales network, if so, taking the enterprise as a weighted average A2 of the first-layer price of the marketing enterprise; if not, then no value is assigned to weighted average A2;
s423, judging whether the number of the enterprises sold by the enterprise at a certain node in the white spirit production enterprise sales network is more than N1If yes, taking the enterprise as a weighted average A3 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then no value is assigned to weighted average A3;
s424, judging whether the number of the enterprises sold by a certain node enterprise in the white spirit production enterprise sales network is more than N2Home and price difference greater than P1If yes, taking the enterprise as a weighted average A4 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then no value is assigned to weighted average A4;
s425, judging whether the selling price of the enterprise at a certain node in the selling network of the white spirit production enterprise is increased more than P2If yes, taking the enterprise as a weighted average A5 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then not to weight the averageAll A5 values are assigned;
s426, judging whether the enterprise is the final layer of the analysis starting point enterprise in the blood relationship network of the white spirit production enterprise, if so, taking the enterprise as a weighted average B1 of one-layer prices of the purchasing parties; if not, no value is assigned to weighted average B1;
s427, judging whether the enterprise is a marketing enterprise with an analysis starting point enterprise as the last layer in the blood margin network of the white spirit production enterprise, if so, taking the enterprise as a weighted average B2 of the first-layer price of the marketing enterprise; if not, no value is assigned to weighted average B2;
s428, whether more than N enterprises are sold by a certain node enterprise in the blood supply network of the white spirit production enterprise or not is judged1If yes, taking the enterprise as a weighted average of the one-level price of the seller B3, and taking the enterprise result closest to the starting point when there are a plurality of such enterprises; if not, no value is assigned to weighted average B3;
s429, judging whether the number of the enterprises sold by a certain node enterprise in the blood relationship network of the white spirit production enterprise is more than N2Home and price difference greater than P1If so, taking the enterprise as a weighted average of the one-tier price of the seller B4, and taking the enterprise result closest to the starting point when there are a plurality of such enterprises; if not, no value is assigned to weighted average B4;
s4210, judging whether the price increase of a certain node enterprise in the blood relationship network of the white spirit production enterprise is more than P2If so, taking the enterprise as a weighted average of the one-tier price of the seller B5, and taking the enterprise result closest to the starting point when there are a plurality of such enterprises; if not, no value is assigned to weighted average B5;
s4211, acquiring a weighted average C of prices sold to retailers by all consanguineous relation enterprises of the white spirit production enterprises in a white spirit retailer sales network;
s4112, when A3 and A4 exist at the same time, retaining the A3 value; when both B3, B4 are present, the B3 value is retained;
s4113, comparing the weighted average results A1, A2, A3, A4 and A5 processed in the step S4112, taking the highest value as a first selling price, and taking the factory price of the selling network of the white spirit production enterprise as the first selling price if all A1, A2, A3, A4 and A5 are not assigned with values; taking the highest value of the weighted average results B1, B2, B3, B4 and B5 as a second selling price, and taking the factory price of the blood relationship network of the white spirit manufacturing enterprise as the second selling price if no value is assigned to B1, B2, B3, B4 and B5; and C is taken as a third selling price, the first selling price, the second selling price and the third selling price are compared, and the highest value is taken as the final selling price.
N1、N2Are integers which are more than or equal to 2 and less than or equal to 10; p1、P2Are all greater than or equal to 0.1 and less than or equal to 1.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.

Claims (7)

1. A risk assessment method for white spirit consumption tax is characterized in that: the method comprises the following steps:
s1, decomposing white spirit elements in the white spirit goods name through white spirit invoice data, and constructing a white spirit dictionary through the white spirit elements;
s2, performing TF/IDF feature extraction on the white spirit elements through the white spirit dictionary, clustering the white spirit elements of all white spirits through a Kmeans clustering algorithm, and extracting all white spirit classes to construct a white spirit class library;
s3, constructing a white spirit tax-related relation network by using tax data;
s4, carrying out statistical analysis on the white spirit tax relationship network, and calculating the delivery price and the sale price of each kind of white spirit;
s5, comparing the calculated delivery price and sale price of each kind of white spirit with the consumption tax declaration data of the kind of white spirit, if the calculated sale price is multiplied by 0.7, the calculated delivery price is larger than the calculated delivery price, and if no product white spirit declaration data exists, the product white spirit is not declared, if the declared delivery price is larger than the declaration lowest tax price or the calculated sale price is multiplied by 0.7, the calculated delivery price is multiplied by 0.6, and the lowest tax price is declared, the product white spirit is false; otherwise, the report is correctly declared.
2. A liquor consumption tax risk assessment method according to claim 1, characterized in that: the white spirit elements comprise name information, specification information and degree information.
3. A liquor consumption tax risk assessment method according to claim 1, characterized in that: the white spirit tax-related relationship network comprises a white spirit production enterprise sales network, a white spirit production enterprise consanguineous network and a white spirit retailer sales network;
the white spirit production enterprise sales network takes white spirit sold by white spirit production enterprises as a starting point, and takes the enterprises as nodes, the purchase-sale relationship as directed edges and the invoice amount and quantity as edge weights according to the types of the white spirit;
the method comprises the steps that a blood margin network of a white spirit production enterprise establishes a non-directional edge with the white spirit production enterprise as an associated blood margin enterprise by taking the white spirit production enterprise as a central node and an enterprise having a common investor relationship with the white spirit production enterprise or an enterprise belonging to the same branch office of the white spirit production enterprise as the associated blood margin enterprise through enterprise common investor and branch office information aiming at the white spirit production enterprise;
the white spirit retailer sales network takes the white spirit purchase of large-scale retailer enterprises as a starting point, the white spirit retailer sales network takes the enterprises as nodes, the purchase-sale relationship is taken as a directed edge, and the amount and quantity of invoices of the white spirit are taken as edge weights.
4. A liquor consumption tax risk assessment method according to claim 3, characterized in that: the factory price calculation method of the white spirit in the step S4 specifically comprises the following steps:
Figure FDA0002708065380000021
wherein P is the factory price of a certain kind of white spirit, N is the total quantity of the kind of white spirit sold by N white spirit production enterprises as sellers, N iskThe amount of white spirit in the kth transaction, UkThe unit price of the white spirit in the kth transaction.
5. A liquor consumption tax risk assessment method according to claim 3, characterized in that: the method for calculating the selling price of the white spirit in the step S4 specifically comprises the following steps:
s421, judging whether the enterprise is the final layer of the analysis starting point enterprise in the white spirit production enterprise sales network, if so, taking the enterprise as a weighted average A1 of a first-layer price of a purchasing party; if not, then no value is assigned to weighted average A1;
s422, judging whether the enterprise is the seller enterprise of the last layer of analysis starting point enterprise in the white spirit production enterprise sales network, if so, taking the enterprise as a weighted average A2 of the first layer of price of the seller; if not, then no value is assigned to weighted average A2;
s423, judging whether the number of the enterprises sold by the enterprise at a certain node in the white spirit production enterprise sales network is more than N1If yes, taking the enterprise as a weighted average A3 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then no value is assigned to weighted average A3;
s424, judging whether more than N enterprises are sold by a certain node enterprise in the white spirit production enterprise sales network2Home and price difference greater than P1If yes, taking the enterprise as a weighted average A4 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then no value is assigned to weighted average A4;
s425, judging whether the selling price of the enterprise at a certain node in the selling network of the white spirit production enterprise is increased by more than P or not2If yes, taking the enterprise as a weighted average A5 of the one-level price of the seller, and taking the enterprise result closest to the starting point when a plurality of such enterprises exist; if not, then no value is assigned to weighted average A5;
s426, judging whether the enterprise is the final layer of the analysis starting point enterprise in the blood relationship network of the white spirit production enterprise, if so, taking the enterprise as a weighted average B1 of one-layer prices of the purchasing party; if not, no value is assigned to weighted average B1;
s427, judging whether the enterprise is the marketing side enterprise of the last layer of the analysis starting point enterprise in the blood margin network of the white spirit production enterprise, if so, taking the enterprise as a weighted average B2 of the first layer of prices of the marketing side; if not, no value is assigned to weighted average B2;
s428, whether more than N enterprises are sold by a certain node enterprise in the blood supply network of the white spirit production enterprise or not is judged1If yes, taking the enterprise as a weighted average of the one-level price of the seller B3, and taking the enterprise result closest to the starting point when there are a plurality of such enterprises; if not, no value is assigned to weighted average B3;
s429, judging whether the number of the enterprises sold by a certain node enterprise in the blood supply network of the white spirit production enterprise is more than N2Home and price difference greater than P1If so, taking the enterprise as a weighted average of the one-tier price of the seller B4, and taking the enterprise result closest to the starting point when there are a plurality of such enterprises; if not, no value is assigned to weighted average B4;
s4210, judging whether the price increase of the enterprise is more than P for a certain node enterprise in the blood relationship network of the white spirit production enterprise2If so, taking the enterprise as a weighted average of the one-tier price of the seller B5, and taking the enterprise result closest to the starting point when there are a plurality of such enterprises; if not, no value is assigned to weighted average B5;
s4211, acquiring a weighted average C of prices sold to retailers by all consanguineous relation enterprises of white spirit production enterprises in the white spirit retailer sales network;
s4112, when A3 and A4 exist at the same time, retaining the A3 value; when both B3, B4 are present, the B3 value is retained;
s4113, comparing the weighted average results A1, A2, A3, A4 and A5 processed in the step S4112, taking the highest value as a first selling price, and taking the factory price of the selling network of the white spirit manufacturing enterprise as the first selling price if none of A1, A2, A3, A4 and A5 is assigned; taking the highest value of the weighted average results B1, B2, B3, B4 and B5 as a second selling price, and taking the factory price of the blood relationship network of the white spirit manufacturing enterprise as the second selling price if no value is assigned to B1, B2, B3, B4 and B5; and C is taken as a third selling price, the first selling price, the second selling price and the third selling price are compared, and the highest value is taken as the final selling price.
6. A liquor consumption tax risk assessment method according to claim 2, characterized in that: the white spirit dictionary is used for extracting white spirit elements from the goods names of the white spirit invoices, and judging whether the goods names belong to white spirit or not according to the white spirit names in the goods names; and matching liquor grades through comparing different liquor factors, and judging whether the two liquor goods names belong to the liquor of the same grade.
7. A liquor consumption tax risk assessment method according to claim 5, characterized in that: said N is1The N2Are integers which are more than or equal to 2 and less than or equal to 10; the P is1The P2Are all greater than or equal to 0.1 and less than or equal to 1.
CN202011046196.XA 2020-09-29 2020-09-29 White spirit consumption tax risk assessment method Pending CN112184022A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011046196.XA CN112184022A (en) 2020-09-29 2020-09-29 White spirit consumption tax risk assessment method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011046196.XA CN112184022A (en) 2020-09-29 2020-09-29 White spirit consumption tax risk assessment method

Publications (1)

Publication Number Publication Date
CN112184022A true CN112184022A (en) 2021-01-05

Family

ID=73945606

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011046196.XA Pending CN112184022A (en) 2020-09-29 2020-09-29 White spirit consumption tax risk assessment method

Country Status (1)

Country Link
CN (1) CN112184022A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105069683A (en) * 2015-07-24 2015-11-18 广州时韵信息科技有限公司 Tax risk assessment system
CN106447476A (en) * 2016-09-08 2017-02-22 广东源恒软件科技有限公司 Automatic check computation method and system of consumer tax
CN108765116A (en) * 2018-05-18 2018-11-06 北京大账房网络科技股份有限公司 Financial intelligence air control method for early warning
CN109271462A (en) * 2018-11-23 2019-01-25 河北航天信息技术有限公司 A kind of taxpayer's tax registration registered address information cluster method based on K-means algorithm model
CN111222971A (en) * 2019-12-31 2020-06-02 航天信息股份有限公司 Method and device for auditing product oil red invoice information table

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105069683A (en) * 2015-07-24 2015-11-18 广州时韵信息科技有限公司 Tax risk assessment system
CN106447476A (en) * 2016-09-08 2017-02-22 广东源恒软件科技有限公司 Automatic check computation method and system of consumer tax
CN108765116A (en) * 2018-05-18 2018-11-06 北京大账房网络科技股份有限公司 Financial intelligence air control method for early warning
CN109271462A (en) * 2018-11-23 2019-01-25 河北航天信息技术有限公司 A kind of taxpayer's tax registration registered address information cluster method based on K-means algorithm model
CN111222971A (en) * 2019-12-31 2020-06-02 航天信息股份有限公司 Method and device for auditing product oil red invoice information table

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李庆云: "Excel支持下的白酒消费税纳税申报风险预警", 财会月刊 *

Similar Documents

Publication Publication Date Title
Lin et al. The effects of supply chain diversification during the COVID-19 crisis: Evidence from Chinese manufacturers
CN110188198B (en) Anti-fraud method and device based on knowledge graph
Dai et al. Industrial clusters, networks and resilience to the Covid-19 shock in China
Yu Processing trade, firm productivity, and tariff reductions: Evidence from Chinese products
Engle et al. Forecasting intraday volatility in the US equity market. Multiplicative component GARCH
US8032448B2 (en) Detecting and measuring risk with predictive models using content mining
CN109711955B (en) Poor evaluation early warning method and system based on current order and blacklist base establishment method
CN112967063B (en) Pond cheat detection method for Ethernet platform
CN106933814A (en) Tax data exception analysis method and system
Schoenle International Menu Costs and Price Dynamics
CN114187084A (en) Method for identifying certificate subjects according to classification of electronic invoice tax
CN114723492A (en) Enterprise portrait generation method and equipment
Aparicio et al. Enhancing the measurement of firm inefficiency accounting for corporate social responsibility: a dynamic data envelopment analysis fuzzy approach
CN110689437A (en) Communication construction project financial risk prediction method based on random forest
CN112419030B (en) Method, system and equipment for evaluating financial fraud risk
CN112184022A (en) White spirit consumption tax risk assessment method
Rybalka The innovative input mix. Assessing the importance of R&D and ICT investments for firm performance in manufacturing and services
CN112465397A (en) Audit data analysis method and device
CN112037006A (en) Credit risk identification method and device for small and micro enterprises
CN115994767A (en) Product supply chain management system based on electronic commerce
Goestjahjanti et al. Significance effect cost of goods sold and inventory on sales PT. Nippon Indosari Corpindo Tbk
CN115439265A (en) Intelligent insurance industry compensation abnormal transaction risk control system
CN113506173A (en) Credit risk assessment method and related equipment thereof
CN115081921A (en) ERP e-commerce management system based on big data
CN114066391A (en) Data processing method and device for pre-qualification and disagreement processing of suppliers

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination