CN109993641A - A kind of invoice writes out falsely method for early warning and system - Google Patents

A kind of invoice writes out falsely method for early warning and system Download PDF

Info

Publication number
CN109993641A
CN109993641A CN201711457960.0A CN201711457960A CN109993641A CN 109993641 A CN109993641 A CN 109993641A CN 201711457960 A CN201711457960 A CN 201711457960A CN 109993641 A CN109993641 A CN 109993641A
Authority
CN
China
Prior art keywords
invoice
information
sales
taxpayer
data
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
CN201711457960.0A
Other languages
Chinese (zh)
Inventor
黄嵘
刘丹
范钢
潘竞旭
谢宇
张玉魁
陈荣兴
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.)
Aisino Corp
Original Assignee
Aisino Corp
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 Aisino Corp filed Critical Aisino Corp
Priority to CN201711457960.0A priority Critical patent/CN109993641A/en
Publication of CN109993641A publication Critical patent/CN109993641A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Abstract

The embodiment of the present invention provides a kind of invoice voiding method for early warning, which is characterized in that this method comprises: the invoice sales information and coupon information according to one or more Target Enterprises establish corresponding dimension table;By the invoice information in sales invoice data set according in preset condition and the dimension table invoice sales information and coupon information be associated and to form result table, and obtain the top volume that invoice amount is greater than threshold value in set period of time in the result table and issue invoice information;Summarize top volume to issue invoice information, and provides according to the top volume summarized the information that issues invoice and write out falsely invoice taxpayer warning information.Provide the tax administration to enterprise for the tax authority, can avoid the occurrence of as far as possible enterprise by write out falsely invoice defrauded the tax, the behavior to tax.

Description

A kind of invoice writes out falsely method for early warning and system
Technical field
The present invention relates to data screening and analyses, write out falsely method for early warning and system more particularly to a kind of invoice.
Background technique
Since 2015, to advanced optimize paying taxes service, tax efficiency is done in raising, and general bureau has carried out " convenience-for-people to do tax spring breeze Action ".The tax authority is carrying out paying taxes service, while tax efficiency is done in raising, should make full use of information-based means, reinforce to increasing It is worth the subsequent monitoring management of tax invoice, trace analysis taxpayer's invoice uses and tax declaration situation.Enterprise is avoided the occurrence of to pass through Write out falsely invoice defrauded the tax, the behavior to tax.
Summary of the invention
The purpose of the embodiment of the present invention is that providing a kind of invoice writes out falsely method for early warning and system, this method by by one or Multiple Target Enterprise sales informations and coupon information and the history invoice data of respective objects enterprise are associated, and acquire association The top volume that invoice amount is greater than threshold value in set period of time in invoice data afterwards issues invoice information, according to the equal letter that issues invoice Breath, which provides, writes out falsely invoice taxpayer warning information.The tax administration to enterprise is provided for the tax authority, can be avoided out as far as possible Existing enterprise defrauded the tax by voiding invoice, the behavior to tax.
To achieve the goals above, the embodiment of the present invention provides a kind of invoice voiding method for early warning, this method comprises:
According to the invoice sales information and coupon information of one or more Target Enterprises, corresponding dimension table is established;
By the invoice information in sales invoice data set according to the invoice sales information in preset condition and the dimension table It is associated to form result table with coupon information, and obtains in the result table invoice amount in set period of time and be greater than threshold value Top volume issues invoice information;
Summarize top volume to issue invoice information, and provides that write out falsely invoice taxpayer pre- according to the top volume summarized the information that issues invoice Alert information.
Optionally, this method further include:
The threshold value is set by following either type:
A default fixed value or the level threshold value that tax authority's offer is provided;
The threshold value is arranged using the method for machine learning.
Optionally, this method further include:
Following operation is executed before establishing the dimension table:
The original invoice data for the one or more Target Enterprises that will acquire import database, form the original hair of Target Enterprise Ticket data collection;And
The sales invoice in one or more Target Enterprise original invoice data in the database is filtered out, institute is provided State sales invoice data set.
Optionally, this method further include: by the sales invoice in the sales invoice data set according to taxpayer's information into Row classification.
Optionally, the preset condition include it is following one or more: invoice codes, invoice number in sales invoice information Code, is made out an invoice the date at Invoice category.
Optionally, this method further include: calculated according to invoicing time and push up volume in set period of time and issue invoice ratio.
Optionally, the set period of time is at least one month.
The embodiment of the present invention also provides a kind of invoice voiding early warning system, which includes:
Calculation control device is used for:
According to the invoice sales information and coupon information of the one or more Target Enterprises obtained by the tax authority, phase is established The dimension table answered;
The invoice information in sales invoice data set that the screening plant is filtered out is according to preset condition and the dimension Invoice sales information and coupon information in degree table are associated, and are obtained of that month invoice amount in relevant invoice and be greater than threshold value Top volume issue invoice information;And
Summarize top volume to issue invoice information, and provides that write out falsely invoice taxpayer pre- according to the top volume summarized the information that issues invoice Alert information.
Optionally, the calculation control device is also used to set the threshold value by following either type:
A default fixed value or the level threshold value that tax authority's offer is provided;
The threshold value is arranged using the method for machine learning.
Optionally, the system also includes:
Database, what the original invoice data for storing the one or more Target Enterprises obtained by the tax authority were established Database forms Target Enterprise original invoice data set;And
Screening plant filters out the pin in one or more Target Enterprise original invoice data in the database purchase Item invoice, provides sales invoice data set.
Optionally, which is also used to the sales invoice in the sales invoice data set according to taxpayer Information is classified.
Optionally, the preset condition include it is following one or more: invoice codes, invoice number in sales invoice information Code, is made out an invoice the date at Invoice category.
Optionally, the calculation control device is also used to: being calculated top volume in set period of time according to invoicing time and is issued Invoice ratio.
Optionally, the set period of time is at least one month.
Through the above technical solutions, issuing invoice information by the top volume for obtaining Target Enterprise in setting time, void is provided Draw a bill taxpayer's warning information, facilitate the tax authority to enterprise's type supervision and management, prevent enterprise by write out falsely invoice into It swindles tax, support tax.
The other feature and advantage of the embodiment of the present invention will the following detailed description will be given in the detailed implementation section.
Detailed description of the invention
Attached drawing is to further understand for providing to the embodiment of the present invention, and constitute part of specification, under The specific embodiment in face is used to explain the present invention embodiment together, but does not constitute the limitation to the embodiment of the present invention.Attached In figure:
Fig. 1 is that a kind of invoice provided in an embodiment of the present invention writes out falsely method for early warning flow diagram;
Fig. 2 is that a kind of invoice provided in an embodiment of the present invention writes out falsely early warning system structural schematic diagram.
Specific embodiment
It is described in detail below in conjunction with specific embodiment of the attached drawing to the embodiment of the present invention.It should be understood that this Locate described specific embodiment and be merely to illustrate and explain the present invention embodiment, is not intended to restrict the invention embodiment.
The noun of locality in embodiments of the present invention, in the absence of explanation to the contrary, used such as " upper and lower, left and right ", " inside and outside ", " remote, close " refer to the direction with reference to attached drawing, and therefore, use direction term is for illustrating not to be to limit this hair It is bright.
Fig. 1 shows a kind of invoice provided in an embodiment of the present invention and writes out falsely method for early warning flow diagram, as shown in Figure 1, This method establishes corresponding dimension table according to the invoice sales information and coupon information of one or more Target Enterprises, and by enterprise Invoice information in the sales invoice data set of target is according to the invoice sales information and ticket letter in preset condition and dimension table Breath is associated to form result table;Wherein, the preset condition can in sales invoice information invoice codes, invoice number, Invoice category, the date of invoice.For example, the parameter in dimension table and the invoice column in sales invoice data set are not associated shape At result table, specifically, choose in the sales invoice information about all invoice informations of corporation tax special invoice and the enterprise It is associated in industry setting time (for example, in some moon) according to the dimension table that sales information is established with coupon information, forms knot Fruit table;And obtain the top volume that invoice amount is greater than threshold value in the setting time from the result table and issue invoice, summarize the enterprise Top volume in the setting time issues invoice, and according to the top volume summarized issue invoice information to the enterprise issue write out falsely invoice Taxpayer's warning information.
In embodiment, before establishing the dimension table, the original invoice number of Target Enterprise can be obtained from the tax authority According to, and the original invoice data of the Target Enterprise are directed into database, the original invoice data set of Target Enterprise is provided.Into one Step ground filters out the sales invoice in database in one or more Target Enterprise original invoice data, provides the pin item hair Ticket data collection.
In embodiment, the sales invoice in sales invoice data set is classified according to taxpayer's information, wherein each Taxpayer is different between Target Enterprise, the sales invoice of multiple enterprises can be classified according to different taxpayer's information.
In embodiment, the time can also be issued according to Target Enterprise sales invoice, calculated in setting time, example Such as, it calculates and provides to push up volume in some month of the enterprise and issue invoice and account for the ratio of the total amount of issuing of this month invoice of the enterprise.
Fig. 2 shows a kind of invoice voiding early warning system structural schematic diagrams provided in an embodiment of the present invention, as shown in Fig. 2, The system may include: calculation control device, be built according to the invoice sales information and coupon information of one or more Target Enterprises Corresponding dimension table is found, and by the invoice information in the sales invoice data set of business goal according in preset condition and dimension table Invoice sales information and coupon information be associated to form result table;Wherein, which can be sales invoice information In invoice codes, invoice number, Invoice category, the date of invoice.For example, by parameter and sales invoice data set in dimension table In invoice column be not associated to form result table, specifically, choose dedicated about corporation tax in the sales invoice information It is established in (for example, in some moon) all invoice informations of invoice and enterprise's setting time according to sales information and coupon information Dimension table is associated, and forms result table;And the top that invoice amount in the setting time is greater than threshold value is obtained from the result table Volume issues invoice, and summarizes top volume of the enterprise in the setting time and issues invoice, and is issued invoice letter according to the top volume summarized It ceases to issue to the enterprise and writes out falsely invoice taxpayer warning information.
In embodiment, which can also include: database and screening plant;Wherein, before establishing the dimension table, The original invoice data of Target Enterprise can be obtained from the tax authority, and the original invoice data of the Target Enterprise are directed into number The original invoice data set of Target Enterprise is provided according to library.Further, it is former that one or more Target Enterprises in database are filtered out The sales invoice in ticket data is originated, the sales invoice data set is provided.
In embodiment, which can also be by the sales invoice in sales invoice data set according to taxpayer Information is classified, wherein taxpayer is different between each Target Enterprise, can be according to different taxpayer's information by multiple enterprises Sales invoice classify.
In embodiment, which can also issue the time according to Target Enterprise sales invoice, calculate In setting time, this month invoice of the enterprise always amount of issuing is accounted for for example, calculating and providing top volume in some moon of the enterprise and issue invoice Ratio.
Embodiment
Obtain the original invoice data of multiple Target Enterprises from the tax authority, and by the original invoice of multiple Target Enterprise Data are directed into database, provide the original invoice data set of Target Enterprise.It will be stored in the database by screening plant Voided check in original invoice data set is filtered, and further filters out sales invoice, provides sales invoice data set. Since the pin side taxpayer of each Target Enterprise is different, by Spark SQL or MapReduce by sales invoice data set according to The taxpayer of each Target Enterprise pin side classifies.Wherein, mode classification is without being limited thereto, can also pass through sales invoice data Administration partition belonging to respective objects enterprise or organ's code is concentrated to classify.Use demand according to user sets class condition Accordingly classified.
The sales information and coupon information that multiple Target Enterprise invoices are obtained from the tax authority, similarly according to each enterprise The sales information that will acquire of pin side taxpayer and coupon information carry out classification processing, and and according to the sales letter after classification processing Breath and coupon information establish corresponding dimension table.
By invoice sales information and ticket in invoice codes in Target Enterprise sales invoice data set and enterprise's setting time Card information is associated to form result table, and specifically, invoice codes include common invoice code and value-added tax according to invoice type Invoice codes, not due to the corresponding meaning of the code length and code of common invoice code and VAT invoice code Together, it can be associated to be formed with coupon information with enterprise's January current year invoice sales information according to the meaning of respective code As a result table.And the enterprise top volume that invoice amount is greater than threshold value within January current year is obtained from the result table and is issued invoice; The time is issued according to enterprise's sales invoice, calculates and top volume in enterprise January current year is provided issues invoice and account for the enterprise The ratio of the total amount of issuing of this month invoice, the top volume for summarizing acquisition issue invoice information, and the top volume information that issues invoice may include The top volume amount of issuing invoice of Target Enterprise January current year and calculated top volume issue invoice ratio;It is opened according to the top volume summarized Have invoice information and issues voiding invoice taxpayer warning information to corresponding enterprise.
For that can be set by two ways with the threshold value being compared that issues invoice in system, two kinds of setting means Including artificial settings and default.In the case where artificial settings, which can preset one for field personnel by rule of thumb Fixed value or the level threshold value that tax authority's offer is provided, wherein the level threshold value that the tax authority provides is that invoice is maximum Issue the 90% of amount.In the case where default, the selected of threshold value can be carried out in a manner of machine learning, especially by Following steps generate and provide default threshold value:
1, the invoice data in former years is acquired as sample space, for example, the previous 3 years invoice datas of acquisition are as sample sky Between;In the sample space 80% data are randomly selected as training set, the data of residue 20% are as test set.
2, confirm and write out falsely invoice taxpayer sample in training set and test set, received invoice is write out falsely in training set and test set Tax people classifies with non-voiding invoice taxpayer.
3, according to write out falsely invoice taxpayer and it is non-write out falsely invoice taxpayer classification setting cost matrix TP, FP, FN, TN }, wherein TP is real class, i.e., example is to write out falsely invoice taxpayer and be predicted as writing out falsely invoice taxpayer;FP is false positive class, I.e. example is to write out falsely invoice taxpayer but be predicted as non-voiding invoice taxpayer;FN is false negative class, i.e. example is non-voiding invoice Taxpayer but be predicted as write out falsely invoice taxpayer;TN is very negative class, i.e., example is non-voiding invoice taxpayer and is predicted as non-void Draw a bill taxpayer.
Table 1 is cost matrix setting
4, it is determined for the cost matrix that invoice data in training set is arranged and invoice is write out falsely to voiding invoice taxpayer and non- The classification accuracy of taxpayer.Specifically, corresponding positive class rate TPR and FPR are calculated separately out, wherein
5, using FPR as abscissa, ROC curve is drawn by ordinate of TPR, AUC is calculated according to the ROC curve of drafting Value, that is, calculate the area under ROC curve, the size be epicycle be calculated estimate threshold value.
6, it is tested in test set.
7, step 1-6 is repeated, training set and test set is randomly selected and carries out more wheels and calculate, calculates and multiple estimates threshold Value, and tested in test set after every wheel calculates.
8, by comparing the calculated result of different rounds, the smallest threshold value of estimating of Select Error is as optimal threshold, i.e., logical More every AUC value taken turns and be calculated is crossed, selects AUC maximum value as optimal threshold.
The system can use Hadoop platform and Spark distributed treatment software frame, facilitate parameter extending transversely, energy Large-scale data calculating is enough carried out, meets the invoice analysis in the whole province even whole nation, realizes the convenience of national invoice on-line analysis, The operating efficiency of tax supervision work is substantially increased, while being strengthened to business taxation tracing management.
The optional embodiment of the embodiment of the present invention is described in detail in conjunction with attached drawing above, still, the embodiment of the present invention is simultaneously The detail being not limited in above embodiment can be to of the invention real in the range of the technology design of the embodiment of the present invention The technical solution for applying example carries out a variety of simple variants, these simple variants belong to the protection scope of the embodiment of the present invention.
It is further to note that specific technical features described in the above specific embodiments, in not lance In the case where shield, it can be combined in any appropriate way.In order to avoid unnecessary repetition, the embodiment of the present invention pair No further explanation will be given for various combinations of possible ways.
It will be appreciated by those skilled in the art that implementing the method for the above embodiments is that can pass through Program is completed to instruct relevant hardware, which is stored in a storage medium, including some instructions are used so that single Piece machine, chip or processor (processor) execute all or part of the steps of each embodiment the method for the application.And it is preceding The storage medium stated includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory The various media that can store program code such as (RAM, Random Access Memory), magnetic or disk.
In addition, any combination can also be carried out between a variety of different embodiments of the embodiment of the present invention, as long as it is not The thought of the embodiment of the present invention is violated, equally should be considered as disclosure of that of the embodiment of the present invention.

Claims (16)

1. a kind of invoice writes out falsely method for early warning, which is characterized in that this method comprises:
According to the invoice sales information and coupon information of one or more Target Enterprises, corresponding dimension table is established;
By the invoice information in sales invoice data set according to the invoice sales information and ticket in preset condition and the dimension table Card information is associated to form result table, and obtains in the result table top volume that invoice amount in set period of time is greater than threshold value Issue invoice information;
Summarize top volume to issue invoice information, and provides according to the top volume summarized the information that issues invoice and write out falsely invoice taxpayer early warning letter Breath.
2. the method according to claim 1, wherein this method further include:
The threshold value is set by following either type:
A default fixed value or the level threshold value that tax authority's offer is provided;
The threshold value is arranged using the method for machine learning.
3. the method according to claim 1, wherein this method further include:
Following operation is executed before establishing the dimension table:
The original invoice data for the one or more Target Enterprises that will acquire import database, form Target Enterprise original invoice number According to collection;And
The sales invoice in one or more Target Enterprise original invoice data in the database is filtered out, the pin is provided Item invoice data collection.
4. the method according to claim 1, wherein this method further include:
Sales invoice in the sales invoice data set is classified according to taxpayer's information.
5. the method according to claim 1, wherein the preset condition include it is following one or more:
Invoice codes, invoice number in sales invoice information, are made out an invoice the date at Invoice category.
6. the method according to claim 1, wherein this method further include:
It is calculated according to invoicing time and pushes up volume in set period of time and issue invoice ratio.
7. according to the method described in claim 6, it is characterized in that, the set period of time is at least one month.
8. according to the method described in claim 2, it is characterized in that, the method for the machine learning comprises the steps of:
S1: the invoice data in acquisition setting time as sample, and carry out in the sample randomly selecting data respectively as Training set and test set;
S2: classify to the voiding invoice taxpayer in training set and test set with non-voiding invoice taxpayer;
S3: according to the classification setting cost matrix { TP, FP, FN, TN } for writing out falsely invoice taxpayer and non-voiding invoice taxpayer, and Determine positive class rate TPR and FPR;
S4: using FPR as abscissa, ROC curve is drawn by ordinate of TPR, AUC value is calculated according to the ROC curve of drafting;
S5: it is tested in test set;
S6: repeatedly setting number step S1-S5;
S7: according to the calculated result of step S6, select AUC maximum value as optimal threshold.
9. a kind of invoice writes out falsely early warning system, which is characterized in that the system includes:
Calculation control device is used for:
According to the invoice sales information and coupon information of the one or more Target Enterprises obtained by the tax authority, establish corresponding Dimension table;
The invoice information in sales invoice data set that the screening plant is filtered out is according to preset condition and the dimension table In invoice sales information and coupon information be associated, and obtain in relevant invoice the top that of that month invoice amount is greater than threshold value Volume issues invoice information;And
Summarize top volume to issue invoice information, and provides according to the top volume summarized the information that issues invoice and write out falsely invoice taxpayer early warning letter Breath.
10. system according to claim 9, which is characterized in that the calculation control device is also used to by following any Mode sets the threshold value:
A default fixed value or the level threshold value that tax authority's offer is provided;
The threshold value is arranged using the method for machine learning.
11. system according to claim 9, which is characterized in that the system also includes:
Database forms the original hair of Target Enterprise for storing the original invoice data of the one or more Target Enterprises obtained Ticket data collection;And
Screening plant filters out the pin item hair in one or more Target Enterprise original invoice data in the database purchase Ticket provides sales invoice data set.
12. system according to claim 9, which is characterized in that the calculation control device is also used to the sales invoice Sales invoice in data set is classified according to taxpayer's information.
13. system according to claim 9, which is characterized in that the preset condition include it is following one or more:
Invoice codes, invoice number in sales invoice information, are made out an invoice the date at Invoice category.
14. system according to claim 9, which is characterized in that the calculation control device is also used to:
It is calculated according to invoicing time and pushes up volume in set period of time and issue invoice ratio.
15. system according to claim 14, which is characterized in that the set period of time is at least one month.
16. system according to claim 10, which is characterized in that the method for the machine learning comprises the steps of:
S1: the invoice data in acquisition setting time as sample, and carry out in the sample randomly selecting data respectively as Training set and test set;
S2: classify to the voiding invoice taxpayer in training set and test set with non-voiding invoice taxpayer;
S3: according to the classification setting cost matrix { TP, FP, FN, TN } for writing out falsely invoice taxpayer and non-voiding invoice taxpayer, and Determine positive class rate TPR and FPR;
S4: using FPR as abscissa, ROC curve is drawn by ordinate of TPR, AUC value is calculated according to the ROC curve of drafting;
S5: it is tested in test set;
S6: repeatedly setting number step S1-S5;
S7: according to the calculated result of step S6, select AUC maximum value as optimal threshold.
CN201711457960.0A 2017-12-28 2017-12-28 A kind of invoice writes out falsely method for early warning and system Pending CN109993641A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711457960.0A CN109993641A (en) 2017-12-28 2017-12-28 A kind of invoice writes out falsely method for early warning and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711457960.0A CN109993641A (en) 2017-12-28 2017-12-28 A kind of invoice writes out falsely method for early warning and system

Publications (1)

Publication Number Publication Date
CN109993641A true CN109993641A (en) 2019-07-09

Family

ID=67108237

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711457960.0A Pending CN109993641A (en) 2017-12-28 2017-12-28 A kind of invoice writes out falsely method for early warning and system

Country Status (1)

Country Link
CN (1) CN109993641A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110599189A (en) * 2019-09-24 2019-12-20 腾讯科技(深圳)有限公司 Bill risk analysis method, related equipment and computer storage medium
CN111222766A (en) * 2019-12-29 2020-06-02 航天信息股份有限公司 Method and system for early warning of enterprise false invoicing
CN111724241A (en) * 2020-06-05 2020-09-29 西安交通大学 Enterprise invoice virtual invoice detection method based on dynamic edge feature enhanced graph attention network
CN112232894A (en) * 2020-10-27 2021-01-15 江苏税软软件科技有限公司 Data analysis method based on value-added tax invoice
CN112633957A (en) * 2020-11-10 2021-04-09 中科金审(北京)科技有限公司 Big data-based false invoice behavior analysis method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103679972A (en) * 2013-12-26 2014-03-26 北京德信易税网络技术有限公司 Monitoring method, device and system of tax-control terminal
CN104700303A (en) * 2013-12-05 2015-06-10 航天信息股份有限公司 Enterprise tax dodge monitoring method through direction analysis of goods flow from added-value tax invoices
CN106780001A (en) * 2016-12-26 2017-05-31 税友软件集团股份有限公司 A kind of invoice writes out falsely enterprise supervision recognition methods and system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104700303A (en) * 2013-12-05 2015-06-10 航天信息股份有限公司 Enterprise tax dodge monitoring method through direction analysis of goods flow from added-value tax invoices
CN103679972A (en) * 2013-12-26 2014-03-26 北京德信易税网络技术有限公司 Monitoring method, device and system of tax-control terminal
CN106780001A (en) * 2016-12-26 2017-05-31 税友软件集团股份有限公司 A kind of invoice writes out falsely enterprise supervision recognition methods and system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
史周华 等: "《中医药统计学与软件应用》", 31 May 2015, 中国中医药出版社 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110599189A (en) * 2019-09-24 2019-12-20 腾讯科技(深圳)有限公司 Bill risk analysis method, related equipment and computer storage medium
CN111222766A (en) * 2019-12-29 2020-06-02 航天信息股份有限公司 Method and system for early warning of enterprise false invoicing
CN111724241A (en) * 2020-06-05 2020-09-29 西安交通大学 Enterprise invoice virtual invoice detection method based on dynamic edge feature enhanced graph attention network
CN111724241B (en) * 2020-06-05 2024-03-29 西安交通大学 Enterprise invoice virtual issuing detection method based on dynamic edge feature graph annotation meaning network
CN112232894A (en) * 2020-10-27 2021-01-15 江苏税软软件科技有限公司 Data analysis method based on value-added tax invoice
CN112633957A (en) * 2020-11-10 2021-04-09 中科金审(北京)科技有限公司 Big data-based false invoice behavior analysis method

Similar Documents

Publication Publication Date Title
CN109993641A (en) A kind of invoice writes out falsely method for early warning and system
US20180374098A1 (en) Modeling method and device for machine learning model
US11734353B2 (en) Multi-sampling model training method and device
KR101597939B1 (en) Apparatus and method for predicting industrial credit risk using macro-economic indicator
Matos et al. An empirical method for discovering tax fraudsters: A real case study of brazilian fiscal evasion
CN109583729B (en) Data processing method and device for platform online model
Fan et al. Improved ML-based technique for credit card scoring in internet financial risk control
CN110634060A (en) User credit risk assessment method, system, device and storage medium
CN109635010A (en) A kind of user characteristics and characterization factor extract, querying method and system
CN109584076A (en) Stock factor income analysis method, apparatus, computer equipment and storage medium
CN110675078A (en) Marketing company risk diagnosis method, system, computer terminal and storage medium
Lukkezen et al. Stochastic debt sustainability indicators 1
CN109583978A (en) The method, device and equipment of invoice enterprise is write out falsely in a kind of identification
CN106875185A (en) A kind of air control model training method and device
CN107679862B (en) Method and device for determining characteristic value of fraud transaction model
CN111046947B (en) Training system and method of classifier and recognition method of abnormal sample
CN109102396A (en) A kind of user credit ranking method, computer equipment and readable medium
CN112365352A (en) Anti-cash-out method and device based on graph neural network
CN112329862A (en) Decision tree-based anti-money laundering method and system
CN112037006A (en) Credit risk identification method and device for small and micro enterprises
CN108229572B (en) Parameter optimization method and computing equipment
US10949863B1 (en) System and method for account abuse risk analysis
CN112884363A (en) Nuclear power project economic evaluation probability risk analysis method
CN113763153A (en) Bank credit risk analysis method and device
CN113191771A (en) Buyer account period risk prediction method

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20190709