CN114092215A - Auditing method and system for export tax refund loan - Google Patents

Auditing method and system for export tax refund loan Download PDF

Info

Publication number
CN114092215A
CN114092215A CN202210062850.9A CN202210062850A CN114092215A CN 114092215 A CN114092215 A CN 114092215A CN 202210062850 A CN202210062850 A CN 202210062850A CN 114092215 A CN114092215 A CN 114092215A
Authority
CN
China
Prior art keywords
transaction
enterprise
declared
tax
rule base
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.)
Granted
Application number
CN202210062850.9A
Other languages
Chinese (zh)
Other versions
CN114092215B (en
Inventor
辛颖梅
朱青
张柳松
罗旻
张子恒
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.)
Nanjing Skytech Quanshuitong Information Technology Co ltd
Original Assignee
Nanjing Skytech Quanshuitong Information Technology 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 Nanjing Skytech Quanshuitong Information Technology Co ltd filed Critical Nanjing Skytech Quanshuitong Information Technology Co ltd
Priority to CN202210062850.9A priority Critical patent/CN114092215B/en
Publication of CN114092215A publication Critical patent/CN114092215A/en
Application granted granted Critical
Publication of CN114092215B publication Critical patent/CN114092215B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/03—Credit; Loans; Processing thereof
    • 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
    • G06Q40/10—Tax strategies

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

The method comprises the steps of obtaining the operation data of an enterprise to be declared and the export data of a project to be declared, generating a rule base for evaluating whether a tax cheating action exists in the enterprise to be declared, grading the tax cheating action on the project to be declared of the enterprise to be declared through the rule base, and judging that the enterprise to be declared is the tax cheating enterprise and does not pass the audit of the tax returned loan if the final grade is higher than a threshold value. The method realizes the correlation analysis among multiple attributes through the method of the incidence relation, can more comprehensively check enterprises applying for handling export tax refunds, and solves the problem that the tax check result obtained only by the customs declaration electronic information and the value-added tax invoice information is not accurate in the prior art.

Description

Auditing method and system for export tax refund loan
Technical Field
The invention relates to the technical field of tax control, in particular to an auditing method and system for export tax refund loan.
Background
The export tax refund loan is a short-term fund difficult problem caused by the fact that export tax refunds of export enterprises cannot be timely paid out by financial institutions such as banks and the like, and is provided for the export enterprises on the premise of trusting the export tax refund accounts of the enterprises, and the export tax refund loan transaction is a short-term mobile fund loan transaction which is guaranteed by the export tax refund accounts. However, while the export tax refund loan business is being developed, the export enterprises who face the business may have fraud, which causes loss to financial institutions such as banks.
In the prior art, manual examination of export enterprises declaring export tax refund loan businesses needs to be performed by means of customs clearance electronic information and value-added tax invoice information, however, the range of tax problems involved is wide, various factors such as export amount, export commodities, export areas and export countries in customs information need to be considered, and the tax examination result obtained by means of the customs clearance electronic information and the value-added tax invoice information is inaccurate.
Disclosure of Invention
The application provides an auditing method and system for export tax refund loan, which aim to solve the problem that in the prior art, the tax auditing result obtained only by means of customs declaration electronic information and value-added tax invoice information is inaccurate.
In one aspect, the application provides a method for auditing an export tax refund loan, comprising:
acquiring operation data of an enterprise to be declared and export data of a project to be declared, wherein the project to be declared is a project which the enterprise to be declared expects to apply for export tax refund loan;
generating a transaction detail set according to the operation data of the enterprise to be declared and the export data of the project to be declared
Figure 253345DEST_PATH_IMAGE001
Said transaction detail set
Figure 345935DEST_PATH_IMAGE001
Including a number of transaction attributes of the enterprise to be declared
Figure 442066DEST_PATH_IMAGE002
Said transaction attribute
Figure 278304DEST_PATH_IMAGE002
Including a number of transaction items
Figure 409071DEST_PATH_IMAGE003
And with said transaction item
Figure 672562DEST_PATH_IMAGE003
Corresponding transaction parameters, wherein,s=1，2，3，……n，i=1，2，3，……n；
according to the transaction item
Figure 990411DEST_PATH_IMAGE003
And the transaction item
Figure 633270DEST_PATH_IMAGE003
Calculating transaction attributes corresponding to the transaction parameters
Figure 618543DEST_PATH_IMAGE002
Concentration ratio of
Figure 928302DEST_PATH_IMAGE004
；
According to the transaction attributes
Figure 858081DEST_PATH_IMAGE002
Concentration ratio of
Figure 911487DEST_PATH_IMAGE004
The transaction attribute is added
Figure 875901DEST_PATH_IMAGE002
Classifying to obtain common transaction item set
Figure 356561DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 39215DEST_PATH_IMAGE006
；
According to the common transaction item set
Figure 630733DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 718163DEST_PATH_IMAGE006
Generating a pseudo rule base
Figure 369724DEST_PATH_IMAGE007
Said pseudo rule base
Figure 149461DEST_PATH_IMAGE007
The enterprise tax cheating risk reporting system is used for judging whether the enterprise to be declared has tax cheating risks or not;
according to the pseudo rule base
Figure 669304DEST_PATH_IMAGE007
Generating a rule base
Figure 483676DEST_PATH_IMAGE008
The rule base is used for scoring the fraud behaviors of the enterprise to be declared;
according to the rule base
Figure 430773DEST_PATH_IMAGE008
Scoring the fraud behaviors of the enterprise to be declared;
according to the rule base
Figure 697806DEST_PATH_IMAGE008
And judging whether the project to be declared passes the audit or not according to the scoring result.
Optionally, the transaction item is based on
Figure 21340DEST_PATH_IMAGE003
And the transaction item
Figure 690219DEST_PATH_IMAGE003
Calculating transaction attributes corresponding to the transaction parameters
Figure 683582DEST_PATH_IMAGE002
Concentration ratio of
Figure 565475DEST_PATH_IMAGE004
The method comprises the following steps:
for each of said transaction attributes separately
Figure 568066DEST_PATH_IMAGE002
The transaction parameters in (1) are sequenced to obtain the transaction attributes
Figure 216085DEST_PATH_IMAGE002
An equal number of transaction parameter ordering sets;
according to the transaction item
Figure 380350DEST_PATH_IMAGE003
Sorting the transaction parameters according to the sequence from big to small according to the sizes of the corresponding transaction parameters to generate a transaction parameter sorting set;
according to
Figure 621976DEST_PATH_IMAGE009
Acquiring the concentration corresponding to the trading parameters in each trading parameter sequencing set
Figure 287313DEST_PATH_IMAGE004
Wherein
Figure 665204DEST_PATH_IMAGE010
is any one of the trade parameters in the sorted set of trade parameters,
Figure 390584DEST_PATH_IMAGE011
is the number of items corresponding to the trading parameters in the trading parameter ordering set,
Figure 853926DEST_PATH_IMAGE012
is that
Figure 198320DEST_PATH_IMAGE011
Is measured.
Optionally, the transaction attribute is determined according to the transaction attributes
Figure 558281DEST_PATH_IMAGE002
Concentration ratio of
Figure 329928DEST_PATH_IMAGE004
The transaction attribute is added
Figure 405201DEST_PATH_IMAGE002
Classifying to obtain common transaction item set
Figure 287706DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 764824DEST_PATH_IMAGE006
The method comprises the following steps:
obtaining the concentration degree in each transaction parameter sequencing set
Figure 707372DEST_PATH_IMAGE004
Maximum value of
Figure 145306DEST_PATH_IMAGE013
The transaction parameter of (a);
obtaining the concentration
Figure 956136DEST_PATH_IMAGE004
Maximum value of
Figure 163127DEST_PATH_IMAGE013
The number of items of the trading parameter arranged in the trading parameter order setx’；
For each of the transaction attributes
Figure 380702DEST_PATH_IMAGE002
The transaction item of
Figure 305933DEST_PATH_IMAGE003
Classifying the transaction attributes
Figure 654875DEST_PATH_IMAGE002
Is located at the firstx’Sum of terms is less thanx’The transaction item corresponding to the item
Figure 981951DEST_PATH_IMAGE003
Is determined as a first set
Figure 723DEST_PATH_IMAGE014
The transaction attribute is added
Figure 537883DEST_PATH_IMAGE002
Is located more than secondx’The transaction item corresponding to the item
Figure 565882DEST_PATH_IMAGE003
Is determined as a second set
Figure 872098DEST_PATH_IMAGE015
；
According to the first set
Figure 327350DEST_PATH_IMAGE014
And the second set
Figure 89158DEST_PATH_IMAGE015
Obtaining the common transaction item set
Figure 655268DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 956937DEST_PATH_IMAGE006
Wherein the common transaction item set
Figure 707724DEST_PATH_IMAGE005
Is the first set
Figure 829264DEST_PATH_IMAGE014
The intersection of items in, the set of uncommon transaction items
Figure 323699DEST_PATH_IMAGE006
For the transaction detail set
Figure 479874DEST_PATH_IMAGE001
With the common transaction item set
Figure 401562DEST_PATH_IMAGE005
The difference of (a).
Optionally, the set of common transaction items is selected from
Figure 10398DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 918311DEST_PATH_IMAGE006
Generating a pseudo rule base
Figure 56556DEST_PATH_IMAGE007
Said pseudo rule base
Figure 24512DEST_PATH_IMAGE007
The method is used for judging whether the enterprise to be declared has a tax fraud risk, and comprises the following steps:
according to the common transaction item set
Figure 245278DEST_PATH_IMAGE005
The transaction item of (1)
Figure 956882DEST_PATH_IMAGE003
Corresponding to the transaction parameters, obtaining the minimum support
Figure 946703DEST_PATH_IMAGE016
Wherein
Figure 85561DEST_PATH_IMAGE017
is the common set of transaction items
Figure 668989DEST_PATH_IMAGE005
The transaction item of (1)
Figure 308918DEST_PATH_IMAGE003
A corresponding minimum value of the transaction parameter;
according to the non-common transaction item set
Figure 28612DEST_PATH_IMAGE006
The transaction item of (1)
Figure 465934DEST_PATH_IMAGE003
Corresponding to the transaction parameters, obtaining the minimum support
Figure 536658DEST_PATH_IMAGE018
Wherein
Figure 714699DEST_PATH_IMAGE019
is the set of uncommon transaction items
Figure 554479DEST_PATH_IMAGE006
The trade item corresponding to the median in (1)
Figure 159772DEST_PATH_IMAGE003
；
Setting a confidence coefficient parameter value and a rule length lower limit parameter value len, and generating the common transaction item set according to an Apriori algorithm
Figure 717793DEST_PATH_IMAGE005
Corresponding first set of rulers
Figure 309311DEST_PATH_IMAGE020
And the set of uncommon transaction items
Figure 128231DEST_PATH_IMAGE006
Corresponding second set of rules
Figure 45372DEST_PATH_IMAGE021
Wherein
Figure 952672DEST_PATH_IMAGE022
for a number of the first set of regulars,
Figure 347882DEST_PATH_IMAGE023
for a number of the second set of regulars,
Figure 286888DEST_PATH_IMAGE024
is (0, 1) type data when
Figure 109350DEST_PATH_IMAGE024
If not less than 0, judging that the enterprise to be declared has no tax fraud risk, and if so, judging that the enterprise to be declared has no tax fraud risk
Figure 501017DEST_PATH_IMAGE024
If =1, determining that the enterprise to be declared has a tax fraud risk;
according to the first rule body set
Figure 699917DEST_PATH_IMAGE025
And the second set of rules
Figure 368796DEST_PATH_IMAGE026
Generating a pseudo rule base
Figure 486794DEST_PATH_IMAGE027
。
Optionally, the rule base is based on the pseudo rule base
Figure 241123DEST_PATH_IMAGE007
Generating a rule base
Figure 105698DEST_PATH_IMAGE008
The method comprises the following steps:
judging whether the enterprise to be declared is a historical tax cheating enterprise or not according to whether the enterprise to be declared has the historical tax cheating behavior or not;
according to the transaction parameters and the pseudo rule base
Figure 629084DEST_PATH_IMAGE007
Each rule body judges whether the enterprise to be declared has a tax fraud risk;
if the target rule body judges that the enterprise to be declared has no tax fraud risk, namely
Figure 183562DEST_PATH_IMAGE028
Then, set the initial value
Figure 425187DEST_PATH_IMAGE029
And calculate
Figure 824944DEST_PATH_IMAGE030
A value of (1), wherein
Figure 468415DEST_PATH_IMAGE031
Is the similarity of the target rule body and the transaction parameter, the target rule body being the pseudo rule base
Figure 803582DEST_PATH_IMAGE007
One of the rulers;
if the target rule body judges that the enterprise to be declared has the tax fraud risk, namely
Figure 657137DEST_PATH_IMAGE032
Then, set the initial value
Figure 735952DEST_PATH_IMAGE033
And calculate
Figure 361493DEST_PATH_IMAGE034
A value of (d);
s11: if the enterprise to be declared does not have historical tax cheating behaviors, judging that the enterprise to be declared is not the historical tax cheating enterprise;
each one obtained by calculation
Figure 133139DEST_PATH_IMAGE030
Value of (A) and
Figure 208412DEST_PATH_IMAGE034
is brought into
Figure 90917DEST_PATH_IMAGE035
To calculate
Figure 568035DEST_PATH_IMAGE036
And is given a value of
Figure 510583DEST_PATH_IMAGE037
=
Figure 948518DEST_PATH_IMAGE038
Wherein
Figure 493768DEST_PATH_IMAGE039
when the enterprise to be declared is not the historical fraud enterprise, the fraud score of the enterprise to be declared,
Figure 966338DEST_PATH_IMAGE036
in order for the parameters to be updated,ta preset fraud score threshold;
according to calculation
Figure 207351DEST_PATH_IMAGE040
Value of (2), update
Figure 132582DEST_PATH_IMAGE037
；
Will be updated
Figure 481524DEST_PATH_IMAGE041
In (1)
Figure 543020DEST_PATH_IMAGE024
To be treated and
Figure 952005DEST_PATH_IMAGE024
is replaced by
Figure 364532DEST_PATH_IMAGE042
Wherein
Figure 392531DEST_PATH_IMAGE042
representing fraud scores of the to-be-declared enterprises;
s12: if the enterprise to be declared has historical tax cheating behaviors, judging that the enterprise to be declared is the historical tax cheating enterprise;
each one obtained by calculation
Figure 964326DEST_PATH_IMAGE030
Value of (A) and
Figure 153999DEST_PATH_IMAGE034
is brought into
Figure 939244DEST_PATH_IMAGE043
To calculate
Figure 505354DEST_PATH_IMAGE040
And is given a value of
Figure 807023DEST_PATH_IMAGE044
=
Figure 292231DEST_PATH_IMAGE045
Wherein
Figure 679350DEST_PATH_IMAGE046
when the enterprise to be declared is the historical fraud enterprise, the fraud score of the enterprise to be declared,
Figure 173785DEST_PATH_IMAGE036
in order for the parameters to be updated,tis a preset fraud score threshold value,
Figure 64381DEST_PATH_IMAGE042
representing fraud scores of the to-be-declared enterprises;
according to calculation
Figure 861435DEST_PATH_IMAGE040
Value of (2), update
Figure 594905DEST_PATH_IMAGE044
；
Will be updated
Figure 768397DEST_PATH_IMAGE047
In (1)
Figure 906642DEST_PATH_IMAGE024
To be treated and
Figure 874598DEST_PATH_IMAGE024
is replaced by
Figure 95364DEST_PATH_IMAGE042
；
Recording
Figure 806968DEST_PATH_IMAGE042
And repeatedly performing S11 or S12, calculating the ratio of the times that the enterprise to be declared is performed S11 or S12gAnd, calculating
Figure 796790DEST_PATH_IMAGE007
In
Figure 935647DEST_PATH_IMAGE024
Is filled in
Figure 909288DEST_PATH_IMAGE037
Or
Figure 159004DEST_PATH_IMAGE044
OfA fraction of a number greater than or equal to 20h；
When in useg>=0.95 andh>=0.99, stop execution of S11 or S12, get rule base
Figure 144277DEST_PATH_IMAGE048
。
Optionally, the above-mentionedg>=0.95 andh>=0.99, stop execution of S11 or S12, get rule base
Figure 316020DEST_PATH_IMAGE048
The method also comprises the following steps:
when in useg>=0.95 andh>=0.99, execution of S11 or S12 is stopped;
deleting
Figure 386744DEST_PATH_IMAGE007
Inlen（
Figure 564785DEST_PATH_IMAGE042
）<10 of
Figure 404565DEST_PATH_IMAGE007
；
Order to
Figure 885225DEST_PATH_IMAGE042
Obtained for the last 20 times
Figure 302300DEST_PATH_IMAGE042
To obtain a rule base
Figure 159397DEST_PATH_IMAGE048
。
Optionally, the rule base
Figure 978317DEST_PATH_IMAGE008
And scoring the fraud behaviors of the enterprise to be declared, and further comprising the following steps:
according to the rule base
Figure 629879DEST_PATH_IMAGE008
Acquiring the information obtained when the enterprise to be declared is the historical tax deception enterprise
Figure 675195DEST_PATH_IMAGE042
Or obtaining the result obtained when the enterprise to be declared is not the historical tax deception enterprise
Figure 197968DEST_PATH_IMAGE042
，
Figure 746761DEST_PATH_IMAGE042
Scoring for fraud.
Optionally, the rule base is used for storing the rule base
Figure 959436DEST_PATH_IMAGE008
Judging whether the item to be declared passes the audit or not according to the scoring result, and further comprising the following steps:
if the fraud score is given
Figure 226470DEST_PATH_IMAGE049
If the enterprise to be declared is a tax cheating enterprise, the enterprise does not pass the audit;
if it is
Figure 550004DEST_PATH_IMAGE050
If the enterprise to be declared is a non-fraud enterprise, the audit is passed.
In another aspect, the present application further provides an auditing system for export tax refunds, an application of the system and the method, where the system includes:
a data sorting module: the method comprises the steps of obtaining operation data of an enterprise to be declared and export data of a project to be declared, wherein the project to be declared is a project which the enterprise to be declared expects to apply for export tax refund loan;
generating a transaction detail set according to the operation data of the enterprise to be declared and the export data of the project to be declared
Figure 953303DEST_PATH_IMAGE001
Said transaction detail set
Figure 212246DEST_PATH_IMAGE001
Including a number of transaction attributes of the enterprise to be declared
Figure 356788DEST_PATH_IMAGE002
Said transaction attribute
Figure 93800DEST_PATH_IMAGE002
Including a number of transaction items
Figure 479170DEST_PATH_IMAGE003
And with said transaction item
Figure 909014DEST_PATH_IMAGE003
Corresponding transaction parameters, wherein,s=1，2，3，……n，i=1，2，3，……n；
a rule generation module: for use in accordance with the transaction item
Figure 885060DEST_PATH_IMAGE003
And the transaction item
Figure 550397DEST_PATH_IMAGE003
Calculating transaction attributes corresponding to the transaction parameters
Figure 193868DEST_PATH_IMAGE002
Concentration ratio of
Figure 653668DEST_PATH_IMAGE004
；
According to the transaction attributes
Figure 382590DEST_PATH_IMAGE002
Concentration of said transaction attributes
Figure 461404DEST_PATH_IMAGE002
Classifying to obtain common transaction item set
Figure 84015DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 590083DEST_PATH_IMAGE006
；
According to the common transaction item set
Figure 933864DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 816370DEST_PATH_IMAGE006
Generating a pseudo rule base
Figure 293487DEST_PATH_IMAGE007
Said pseudo rule base
Figure 360669DEST_PATH_IMAGE007
The method is used for judging whether the enterprise to be declared has a tax cheating risk or not;
according to the pseudo rule base
Figure 798604DEST_PATH_IMAGE007
Generating a rule base
Figure 219221DEST_PATH_IMAGE008
The rule base is used for scoring the fraud behaviors of the enterprise to be declared;
an auditing module: for use in accordance with the rule base
Figure 82004DEST_PATH_IMAGE008
Scoring the fraud behaviors of the enterprise to be declared;
according to the rule base
Figure 929874DEST_PATH_IMAGE008
And judging whether the project to be declared passes the audit or not according to the scoring result.
Optionally, the system further comprises a database module, a self-learning module and a rule base module,
the database module is used for storing case samples, and the case samples comprise operation data of historical reporting enterprises, export data of reporting projects and auditing results of the reporting projects;
the self-learning module is used for indicating the data sorting module and the rule generating module to calculate the pseudo rule base corresponding to the case samples according to the case samples stored in the database module at regular intervals
Figure 717089DEST_PATH_IMAGE007
And said rule base
Figure 941397DEST_PATH_IMAGE008
；
The rule base module is used for storing the pseudo rule base which is generated latest according to the indication of the self-learning module
Figure 393107DEST_PATH_IMAGE007
And said rule base
Figure 677457DEST_PATH_IMAGE008
。
It should be noted that, based on the above description of the method and system provided by the present application, the beneficial effects of the present application are as follows:
(1) the present application creates parameters
Figure 214618DEST_PATH_IMAGE051
The method for calculating the concentration of the current attribute can well distinguish a large number of samples into common samples and uncommon samples and respectively generate rules.
(2) The application provides the limitation of the rule length, so that the interference of a large number of rules with different lengths on the result is avoided in the using process.
(3) The application provides a method for calculating the support degree, and the support degree
Figure 977038DEST_PATH_IMAGE052
Which isThe result is a dynamic value, dependent on the argument
Figure 158620DEST_PATH_IMAGE010
The support degree is set to be a constant parameter, and the obtained result is more accurate.
(4) The method realizes tax cheating auditing based on a mode of learning the rules of the enterprises judged to be tax cheating, namely the method has self-learning capability.
(5) The application can automatically score the fraud behaviors of the enterprise to be declared, thereby not only improving the interpretation capability of the system on the result, but also reducing the use threshold of the system;
(6) the application uses tax auditing facing to export tax refund, is suitable for various cases with complex auditing attributes, depends on more enterprise operation data rather than pure tax data, and further raises the threshold of tax counterfeiting of export enterprises, so that the tax auditing data is more accurate.
According to the technical scheme, the method can be realized through the system, the method generates a rule base for evaluating whether the enterprise to be declared has tax cheating behaviors or not by acquiring the operation data of the enterprise to be declared and the export data of the project to be declared, scores the tax cheating behaviors of the project to be declared of the enterprise to be declared through the rule base, and judges that the enterprise to be declared is a tax cheating enterprise and does not pass the audit of the tax returned loan if the final score is higher than a threshold value. The method realizes the correlation analysis among multiple attributes through the method of the incidence relation, can more comprehensively check enterprises applying for handling export tax refunds, and solves the problem that the tax check result obtained only by the customs declaration electronic information and the value-added tax invoice information is not accurate in the prior art.
Drawings
FIG. 1 is a flow chart of an audit method for an export tax refund provided by the present application;
fig. 2 is a block diagram of an audit system for export tax refunds according to the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly described below with reference to the drawings in the embodiments of the present application. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all embodiments. Other embodiments based on the embodiments of the present application and obtained by a person of ordinary skill in the art without any creative effort belong to the protection scope of the present application.
To make the purpose and embodiments of the present application clearer, the following will clearly and completely describe the exemplary embodiments of the present application with reference to the attached drawings in the exemplary embodiments of the present application, and it is obvious that the described exemplary embodiments are only a part of the embodiments of the present application, and not all of the embodiments.
It should be noted that the brief descriptions of the terms in the present application are only for the convenience of understanding the embodiments described below, and are not intended to limit the embodiments of the present application. These terms should be understood in their ordinary and customary meaning unless otherwise indicated.
The terms "first," "second," "third," and the like in the description and claims of this application and in the above-described drawings are used for distinguishing between similar or analogous objects or entities and not necessarily for describing a particular sequential or chronological order, unless otherwise indicated. It is to be understood that the terms so used are interchangeable under appropriate circumstances.
The terms "comprises" and "comprising," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a product or apparatus that comprises a list of elements is not necessarily limited to all elements expressly listed, but may include other elements not expressly listed or inherent to such product or apparatus.
The term "module" refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and/or software code that is capable of performing the functionality associated with that element.
The export tax refund loan is short for export tax refund account escrow loan, and the export tax refund loan enables export enterprises to obtain mobile funds for turnover in advance from financial institutions such as banks and the like under the condition that the tax refund is not up, thereby not only ensuring the virtuous circle of the export business of the enterprises, but also supporting the expansion of the export of the enterprises to a certain extent. For export foreign trade enterprises, the characteristics of digitalization, standardization and centralization are increasingly obvious under strong supervision of customs, outer tubes, tax affairs and commerce. For tax authorities, whether the export growth trend is reasonable or not, whether the operated commodities are reasonable or not and whether the upstream and downstream stability is reliable or not are longitudinally analyzed through electronic data of the foreign trade operation of enterprises, and then the future operation trend of the foreign trade enterprises is predicted through the enterprise operation data of the same industry transversely. Therefore, banks and other financial institutions hope that a method can be used for carrying out correlation analysis on multi-dimensional information of export enterprises, on one hand, distinguishing tax refunds of different enterprises is distinguished, so that distinguishing analysis is carried out, on the other hand, analysis can be carried out according to the characteristics of business behaviors of enterprises judged to have tax cheating behaviors in the past, the characteristics of the tax cheating behaviors are mined in a machine learning mode, the enterprise behaviors with tax cheating doubts are screened in batches, and therefore intelligent and personalized targeted screening of the business behaviors of mass export enterprises is achieved.
On the basis, the method and the system for auditing the export tax refund loan are provided by the application, the behavior is summarized and summarized based on the discovered cheating behavior of the cheating enterprise, the cheating behavior of the enterprise reporting the export tax refund loan is automatically scored to judge whether the enterprise is the cheating enterprise, the workload of cheating audit in the process of auditing the mortgages of other financial institutions such as banks is greatly reduced, and the multi-index combined analysis is realized aiming at the characteristic that the special related factors of an export tax refund link are too many.
Fig. 1 is a flowchart of an auditing method for export tax refunds provided by the present application, and as shown in fig. 1, the auditing method for export tax refunds provided by the present application includes:
s100: the method comprises the steps of obtaining operation data of an enterprise to be declared and export data of a project to be declared, wherein the project to be declared is a project of the enterprise to be declared, which expects to apply for export tax refund.
In some embodiments, the business data of the enterprise to be declared may include financial conditions of the enterprise to be declared, historical loan application data, whether tax cheating behaviors exist or not, and the like, and the export data of the project to be declared may include export tax declaration data, transaction document quantity data, customs declaration data, logistics data, customs verification data and the like of the project for which the enterprise to be declared expects to apply for export tax refund loan at this time.
S110: generating a transaction detail set according to the operation data of the enterprise to be declared and the export data of the project to be declared
Figure 738506DEST_PATH_IMAGE001
Transaction detail collection
Figure 638329DEST_PATH_IMAGE001
Including transaction attributes of a business to be declared
Figure 597582DEST_PATH_IMAGE002
Transaction attribute
Figure 633671DEST_PATH_IMAGE002
Including a number of transaction items
Figure 259825DEST_PATH_IMAGE003
And with transaction items
Figure 505998DEST_PATH_IMAGE003
Corresponding transaction parameters, wherein,s=1，2，3，……n，i=1，2，3，……n。
in some embodiments, the transaction detail set
Figure 610221DEST_PATH_IMAGE053
Wherein
Figure 891029DEST_PATH_IMAGE054
represents a target value when
Figure 688084DEST_PATH_IMAGE054
When =0, the enterprise to be declared is a non-fraud enterprise, and when the enterprise to be declared is a non-fraud enterprise
Figure 687133DEST_PATH_IMAGE054
When =1, the enterprise to be declared is a tax cheating enterprise,
Figure 860625DEST_PATH_IMAGE055
representssThe transaction attributes, for example,
Figure 709853DEST_PATH_IMAGE056
representing the export country,
Figure 677809DEST_PATH_IMAGE057
Representing an export continent,
Figure 773941DEST_PATH_IMAGE058
Representing the average value of the growth rate of the exported month, etc., then
Figure 344600DEST_PATH_IMAGE059
Wherein, export country, export continent and export month growth rate mean value are transaction attributes, each of which comprises several transaction items
Figure 475367DEST_PATH_IMAGE003
And with said transaction item
Figure 738858DEST_PATH_IMAGE003
Corresponding transaction parameters, e.g., export countries may include country A, B, and C, etc., then
Figure 322286DEST_PATH_IMAGE056
= (nation A, nation B, nation C)
Figure 696636DEST_PATH_IMAGE060
Country N), for countries a, B, C
Figure 681909DEST_PATH_IMAGE060
For each transaction item of the N countries, the corresponding parameter may be the transaction amount between the enterprise to be declared and each country, or the number of documents exported to each country by the enterprise to be declared.
S120: according to the transaction item
Figure 384811DEST_PATH_IMAGE003
And transaction items
Figure 455535DEST_PATH_IMAGE003
Calculating transaction attributes corresponding to the transaction parameters
Figure 508941DEST_PATH_IMAGE002
Concentration ratio of
Figure 207776DEST_PATH_IMAGE004
。
In some embodiments, the transaction attributes are individually for each of the transaction attributes
Figure 688436DEST_PATH_IMAGE002
The transaction parameters in (1) are sorted to obtain the transaction attributes
Figure 105511DEST_PATH_IMAGE002
An equal number of transaction parameter ordering sets for trading items
Figure 962608DEST_PATH_IMAGE003
And sequencing the corresponding transaction parameters from big to small to generate a transaction parameter sequencing set.
For example,
Figure 47108DEST_PATH_IMAGE056
= (country a, country B, country C, country D, country E), wherein the transaction parameters corresponding to the transaction items of country a, country B, country C, country D, country E are export receipt number, and country a is rightThe number of the export documents is 1000, the number of the export documents corresponding to the country B is 500, the number of the export documents corresponding to the country C is 1500, the number of the export documents corresponding to the country D is 1700, the number of the export documents corresponding to the country E is 700, and a transaction parameter sorting set is generated according to the number of the export documents corresponding to the transaction items of the countries A, B, C, D and E
Figure 698669DEST_PATH_IMAGE061
Wherein
Figure 605970DEST_PATH_IMAGE062
=1700，
Figure 1179DEST_PATH_IMAGE062
expressed as the number of export documents corresponding to country D,
Figure 549972DEST_PATH_IMAGE063
=1500，
Figure 762648DEST_PATH_IMAGE063
expressed as the number corresponding to country C,
Figure 29681DEST_PATH_IMAGE064
=1000，
Figure 87636DEST_PATH_IMAGE064
expressed as the number of export documents corresponding to country a,
Figure 756514DEST_PATH_IMAGE065
=700，
Figure 140091DEST_PATH_IMAGE065
expressed as the number of export documents corresponding to country E,
Figure 628841DEST_PATH_IMAGE066
=500，
Figure 24575DEST_PATH_IMAGE066
and the number of export documents corresponding to the country B is expressed.
Figure 547960DEST_PATH_IMAGE067
、
Figure 712225DEST_PATH_IMAGE068
、
Figure 812905DEST_PATH_IMAGE069
、
Figure 353608DEST_PATH_IMAGE070
、
Figure 856134DEST_PATH_IMAGE071
Respectively represent the corresponding
Figure 456879DEST_PATH_IMAGE010
The number of terms of the function in the transaction parameter ordering set, wherein,
Figure 310435DEST_PATH_IMAGE067
=1，
Figure 654828DEST_PATH_IMAGE067
=2，
Figure 152806DEST_PATH_IMAGE069
=3，
Figure 520858DEST_PATH_IMAGE070
=4，
Figure 737075DEST_PATH_IMAGE071
=5。
in some embodiments, according to
Figure 744215DEST_PATH_IMAGE072
The concentration degree corresponding to the trading parameters in each trading parameter sequencing set can be obtained
Figure 96698DEST_PATH_IMAGE004
Wherein
Figure 898301DEST_PATH_IMAGE010
is any one of the trade parameters in the sorted set of trade parameters,
Figure 336236DEST_PATH_IMAGE011
is the number of items corresponding to the trading parameters in the trading parameter ordering set,
Figure 147066DEST_PATH_IMAGE012
is that
Figure 354056DEST_PATH_IMAGE011
Is measured. In transaction detail collection
Figure 860649DEST_PATH_IMAGE053
Any one transaction attribute
Figure 785879DEST_PATH_IMAGE002
Correspond to the transaction items with which they are included
Figure 10187DEST_PATH_IMAGE003
Concentration of equal amount
Figure 196318DEST_PATH_IMAGE004
。
S130: according to each transaction attribute
Figure 480669DEST_PATH_IMAGE002
Concentration ratio of
Figure 752250DEST_PATH_IMAGE004
Attributes of the transaction
Figure 780249DEST_PATH_IMAGE002
Classifying to obtain common transaction item set
Figure 352045DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 807297DEST_PATH_IMAGE006
。
In some embodiments, a concentration of each ranked set of transaction parameters is obtained
Figure 707120DEST_PATH_IMAGE004
Maximum value of
Figure 135214DEST_PATH_IMAGE013
And, obtaining a transaction parameter having a concentration
Figure 171303DEST_PATH_IMAGE004
Maximum value of
Figure 797457DEST_PATH_IMAGE013
The number of items of the trading parameter arranged in a trading parameter ordering setx’For each transaction attribute
Figure 309210DEST_PATH_IMAGE002
Transaction item of
Figure 679011DEST_PATH_IMAGE003
Classifying the transaction attributes
Figure 694241DEST_PATH_IMAGE002
Is located at the firstx’Sum of terms is less thanx’Transaction item corresponding to item
Figure 491295DEST_PATH_IMAGE003
Is determined as a first set
Figure 224765DEST_PATH_IMAGE014
Attributes of the transaction
Figure 132678DEST_PATH_IMAGE002
Is located more than secondx’Transaction item corresponding to item
Figure 536502DEST_PATH_IMAGE003
Is determined by the setSet as the second set
Figure 504458DEST_PATH_IMAGE015
。
For example, in
Figure 600590DEST_PATH_IMAGE061
In, if
Figure 171249DEST_PATH_IMAGE064
Corresponds to having a maximum value
Figure 302016DEST_PATH_IMAGE013
Concentration ratio of
Figure 565507DEST_PATH_IMAGE004
Then obtain
Figure 883356DEST_PATH_IMAGE064
Ordering sets in transaction parameters
Figure 398651DEST_PATH_IMAGE073
Number of items in (1) 3, andx’=3. for each transaction attribute
Figure 774137DEST_PATH_IMAGE002
Transaction item of
Figure 83896DEST_PATH_IMAGE003
Classifying the transaction attributes
Figure 774463DEST_PATH_IMAGE002
The transaction items corresponding to items 1, 2 and 3
Figure 827869DEST_PATH_IMAGE003
Is determined as a first set
Figure 667649DEST_PATH_IMAGE014
Attributes of the transaction
Figure 538522DEST_PATH_IMAGE002
The transaction items corresponding to items 4 and 5 in the list
Figure 96542DEST_PATH_IMAGE003
Is determined as a second set
Figure 547115DEST_PATH_IMAGE015
. When in use
Figure 506981DEST_PATH_IMAGE056
=（
Figure 283176DEST_PATH_IMAGE074
) Time, first set
Figure 62913DEST_PATH_IMAGE075
Second set of
Figure 458122DEST_PATH_IMAGE015
=（
Figure 665637DEST_PATH_IMAGE076
）。
In some embodiments, according to the first set
Figure 488100DEST_PATH_IMAGE014
And a second set
Figure 614188DEST_PATH_IMAGE015
Common transaction item sets can be obtained
Figure 813088DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 481967DEST_PATH_IMAGE006
Wherein a common set of transaction items
Figure 599964DEST_PATH_IMAGE005
Is a first set
Figure 354294DEST_PATH_IMAGE014
The intersection of the items in the document, the unusual transaction item set
Figure 747098DEST_PATH_IMAGE006
For the transaction detail set
Figure 270483DEST_PATH_IMAGE001
With the common transaction item set
Figure 296732DEST_PATH_IMAGE005
The difference of (a).
More specifically, transaction detail sets
Figure 538358DEST_PATH_IMAGE077
When it is in the first set
Figure 79061DEST_PATH_IMAGE075
Second set of
Figure 847165DEST_PATH_IMAGE015
=（
Figure 447911DEST_PATH_IMAGE076
) Occasionally, a common set of transaction items
Figure 770308DEST_PATH_IMAGE078
Rare transaction item set
Figure 114702DEST_PATH_IMAGE006
=
Figure 347100DEST_PATH_IMAGE079
。
S140: according to common transaction item set
Figure 243380DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 194019DEST_PATH_IMAGE006
Generating a pseudo rule base
Figure 469667DEST_PATH_IMAGE007
Pseudo rule base
Figure 822151DEST_PATH_IMAGE007
The method is used for judging whether the enterprise to be declared has the tax fraud risk.
In some embodiments, the set of common transaction items is based on
Figure 764699DEST_PATH_IMAGE005
The transaction item in (1)
Figure 61688DEST_PATH_IMAGE003
Corresponding transaction parameters can obtain common transaction item sets
Figure 747885DEST_PATH_IMAGE005
The minimum degree of support in (2) is,
Figure 79509DEST_PATH_IMAGE080
wherein
Figure 192958DEST_PATH_IMAGE017
is a common set of transaction items
Figure 508402DEST_PATH_IMAGE005
The transaction item in (1)
Figure 732710DEST_PATH_IMAGE003
A minimum value of the corresponding transaction parameter; based on an uncommon transaction item set
Figure 59786DEST_PATH_IMAGE006
The transaction item in (1)
Figure 940542DEST_PATH_IMAGE003
Corresponding transaction parameters can obtain a set of uncommon transaction items
Figure 353069DEST_PATH_IMAGE006
Minimum support degree in (1)
Figure 771281DEST_PATH_IMAGE081
Wherein
Figure 952863DEST_PATH_IMAGE019
is the set of uncommon transaction items
Figure 267170DEST_PATH_IMAGE006
The trade item corresponding to the median in (1)
Figure 901414DEST_PATH_IMAGE003
。
Further, a confidence parameter value is setconfidenceAnd a value of a lower limit parameter of the regular lengthlenAccording to the Apriori algorithm, a set of common transaction items can be generated
Figure 467524DEST_PATH_IMAGE005
Corresponding first set of rulers
Figure 893826DEST_PATH_IMAGE020
And an uncommon transaction item set
Figure 519980DEST_PATH_IMAGE006
Corresponding second set of rules
Figure 34662DEST_PATH_IMAGE021
Wherein
Figure 404464DEST_PATH_IMAGE082
are a number of the first set of regulars,
Figure 419693DEST_PATH_IMAGE023
for a number of the second set of regulars,
Figure 216748DEST_PATH_IMAGE024
is (0, 1) type data when
Figure 950217DEST_PATH_IMAGE024
If not less than 0, the enterprise to be declared is judged not to have the risk of fraud, and if not, the enterprise to be declared is judged to have the risk of fraud
Figure 858130DEST_PATH_IMAGE024
And when the statement value is not less than 1, judging that the enterprise to be declared has a tax fraud risk.
The support degree is a parameter indicating a ratio of records in a certain data set that include a certain item set, that is, a frequency of occurrence of a certain item set in the data set, and is used to measure a frequency of the item set. Confidence is defined for association rules, representing the probability of a set of items appearing under specified conditions, to measure the relationship between sets of items. The Apriori algorithm is an association rule mining algorithm, which uses an iterative method of layer-by-layer search to find out the relationship of item sets in a database to form a rule, and the process of the algorithm consists of connection (class matrix operation) and pruning (removing unnecessary intermediate results). The concept of a set of terms in the algorithm is a set of terms. ComprisesKThe set of items iskA set of items. The frequency of occurrence of a set of items is the number of transactions that contain the set of items, referred to as the frequency of the set of items. If a certain item set meets the minimum support, it is called a frequent item set.
In some embodiments, the set of rules is based on a first set of rules
Figure 868812DEST_PATH_IMAGE025
And a second set of rules
Figure 226981DEST_PATH_IMAGE026
A pseudo rule base can be generated
Figure 323113DEST_PATH_IMAGE027
。
S150: according to the pseudo rule base
Figure 896701DEST_PATH_IMAGE007
Generating a rule base
Figure 761889DEST_PATH_IMAGE008
And the rule base is used for scoring the fraud behaviors of the enterprise to be declared.
In some embodiments, whether the enterprise to be declared is a historical tax cheating enterprise can be judged according to whether the enterprise to be declared has the historical tax cheating behavior;
according to the transaction parameters and the pseudo rule base
Figure 900746DEST_PATH_IMAGE007
Each rule body judges whether the enterprise to be declared has a tax fraud risk;
if the target rule body judges that the enterprise to be declared has no tax fraud risk, namely
Figure 874387DEST_PATH_IMAGE028
Then, set the initial value
Figure 389682DEST_PATH_IMAGE029
And calculate
Figure 968431DEST_PATH_IMAGE030
A value of (1), wherein
Figure 278190DEST_PATH_IMAGE031
Is the similarity of the target rule body and the transaction parameter, the target rule body being the pseudo rule base
Figure 473548DEST_PATH_IMAGE007
One of the rulers;
if the target rule body judges that the enterprise to be declared has the tax fraud risk, namely
Figure 526954DEST_PATH_IMAGE083
Then, set the initial value
Figure 366734DEST_PATH_IMAGE033
And calculate
Figure 240537DEST_PATH_IMAGE034
A value of (d);
in some embodiments, each rule body is calculated to correspond to
Figure 798557DEST_PATH_IMAGE030
Value of or
Figure 249130DEST_PATH_IMAGE034
After the value of (3), depending on whether the enterprise to be declared is a historical tax fraudster, the step S11 or S12 may be executed:
s11: if the enterprise to be declared does not have the historical tax cheating behavior, judging that the enterprise to be declared is not the historical tax cheating enterprise;
each one obtained by calculation
Figure 943417DEST_PATH_IMAGE030
Value of (A) and
Figure 250770DEST_PATH_IMAGE034
is brought into
Figure 30507DEST_PATH_IMAGE035
To calculate
Figure 425717DEST_PATH_IMAGE036
And is given a value of
Figure 364723DEST_PATH_IMAGE037
=
Figure 187185DEST_PATH_IMAGE038
Wherein
Figure 292765DEST_PATH_IMAGE039
when the enterprise to be declared is not the historical fraud enterprise, the fraud score of the enterprise to be declared,
Figure 491665DEST_PATH_IMAGE036
in order for the parameters to be updated,ta preset fraud score threshold;
according to calculation
Figure 550757DEST_PATH_IMAGE040
Value of (2), update
Figure 544121DEST_PATH_IMAGE037
；
Will be updated
Figure 157505DEST_PATH_IMAGE041
In (1)
Figure 894517DEST_PATH_IMAGE024
To be treated and
Figure 417902DEST_PATH_IMAGE024
is replaced by
Figure 972380DEST_PATH_IMAGE042
Wherein
Figure 214006DEST_PATH_IMAGE042
representing fraud scores of the to-be-declared enterprises;
s12: if the enterprise to be declared has historical tax cheating behaviors, judging that the enterprise to be declared is the historical tax cheating enterprise;
each one obtained by calculation
Figure 882272DEST_PATH_IMAGE030
Value of (A) and
Figure 525743DEST_PATH_IMAGE034
is brought into
Figure 860909DEST_PATH_IMAGE043
To calculate
Figure 448885DEST_PATH_IMAGE040
And is given a value of
Figure 527700DEST_PATH_IMAGE044
=
Figure 415890DEST_PATH_IMAGE045
Wherein
Figure 187537DEST_PATH_IMAGE046
when the enterprise to be declared is the historical fraud enterprise, the fraud score of the enterprise to be declared,
Figure 262809DEST_PATH_IMAGE036
in order for the parameters to be updated,tis a preset fraud score threshold value,
Figure 145315DEST_PATH_IMAGE042
representing fraud scores of the to-be-declared enterprises;
according to calculation
Figure 497799DEST_PATH_IMAGE040
Value of (2), update
Figure 302331DEST_PATH_IMAGE044
；
Will be updated
Figure 740266DEST_PATH_IMAGE047
In (1)
Figure 551096DEST_PATH_IMAGE024
To be treated and
Figure 23665DEST_PATH_IMAGE024
is replaced by
Figure 996169DEST_PATH_IMAGE042
；
Recording
Figure 921400DEST_PATH_IMAGE042
And repeatedly performing S11 or S12, calculating the ratio of the times that the enterprise to be declared is performed S11 or S12gAnd, calculating
Figure 270342DEST_PATH_IMAGE007
In
Figure 331839DEST_PATH_IMAGE024
Is filled in
Figure 616190DEST_PATH_IMAGE037
Or
Figure 421859DEST_PATH_IMAGE044
The number of times of (2) is greater than or equal to 20h。
When in useg>=0.95 andh>=0.99, stop execution of S11 or S12, get rule base
Figure 449858DEST_PATH_IMAGE048
。
Wherein,
Figure 490495DEST_PATH_IMAGE042
is the assignment data of the number of the points,
Figure 680168DEST_PATH_IMAGE084
and the rule body is used for scoring the input cheating behavior of the enterprise to be declared.
In some embodiments, deletion may be made
Figure 970204DEST_PATH_IMAGE007
Inlen（
Figure 536314DEST_PATH_IMAGE042
）<10 of
Figure 962617DEST_PATH_IMAGE007
Let us order
Figure 323191DEST_PATH_IMAGE042
Obtained for the last 20 times
Figure 572294DEST_PATH_IMAGE042
To obtain a rule base
Figure 942096DEST_PATH_IMAGE048
。
S160: according to the rule base
Figure 832691DEST_PATH_IMAGE008
And scoring the fraud behaviors of the enterprise to be declared.
In some embodiments, based on the resulting rule base
Figure 19959DEST_PATH_IMAGE008
The information obtained when the enterprise to be declared is a historical tax cheating enterprise can be obtained
Figure 628795DEST_PATH_IMAGE042
Or obtained when the enterprise to be declared is not a historical tax cheating enterprise
Figure 661342DEST_PATH_IMAGE042
，
Figure 672023DEST_PATH_IMAGE042
For the fraud score of the enterprise to be declared,
Figure 764613DEST_PATH_IMAGE042
the higher the score of (a) is, the more fraud is indicated to exist in the enterprise to be declared.
S170: according to a rule base
Figure 860745DEST_PATH_IMAGE008
And judging whether the item to be declared passes the audit or not according to the scoring result.
In some embodiments, a fraud threshold t may be preset as a criterion for evaluating whether the enterprise to be declared is a fraud enterprise and whether the item to be declared passes the examination of the export tax refund, that is, if the fraud score of the enterprise to be declared
Figure 965492DEST_PATH_IMAGE049
If the enterprise to be declared is a fraud enterprise, the tax fraud is not checked, and if the fraud score of the enterprise to be declared is the fraud score
Figure 830679DEST_PATH_IMAGE085
And if the enterprise to be declared is judged to be a non-tax cheating enterprise, the enterprise is approved to handle tax refunding loan.
According to the auditing method of the export tax refund loan, the application also provides an auditing system of the export tax refund loan, and the structure diagram of the auditing system of the export tax refund loan provided by the application is shown in fig. 2, and the system comprises a data sorting module, a responsible generation module, an auditing module, a database module, a self-learning module and a rule base module, wherein the data sorting module is connected with the rule generation module in a one-way mode, the rule generation module is connected with the auditing module in a one-way mode, and the rule base module and the database module are respectively connected with the rule generation module in two-way mode in the self-learning module.
In some embodiments, the data arrangement module is used for acquiring operation data of an enterprise to be declared and export data of a project to be declared, wherein the project to be declared is a project for which the enterprise to be declared expects to apply for export tax refund;
generating a transaction detail set according to the operation data of the enterprise to be declared and the export data of the project to be declared
Figure 828591DEST_PATH_IMAGE001
Said transaction detail set
Figure 412019DEST_PATH_IMAGE001
Including a number of transaction attributes of the enterprise to be declared
Figure 51948DEST_PATH_IMAGE002
Said transaction attribute
Figure 37222DEST_PATH_IMAGE002
Including a number of transaction items
Figure 471614DEST_PATH_IMAGE003
And with said transaction item
Figure 542338DEST_PATH_IMAGE003
Corresponding transaction parameters, wherein,s=1，2，3，……n，i=1，2，3，……n；
a rule generation module: for use in accordance with the transaction item
Figure 457729DEST_PATH_IMAGE003
And the transaction item
Figure 297509DEST_PATH_IMAGE003
Calculating transaction attributes corresponding to the transaction parameters
Figure 168382DEST_PATH_IMAGE002
Concentration ratio of
Figure 460823DEST_PATH_IMAGE004
；
According to the transaction attributes
Figure 317921DEST_PATH_IMAGE002
Concentration ratio of
Figure 871262DEST_PATH_IMAGE004
The transaction attribute is added
Figure 522823DEST_PATH_IMAGE002
Classifying to obtain common transaction item set
Figure 958352DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 353562DEST_PATH_IMAGE006
；
According to the common transaction item set
Figure 29918DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 117960DEST_PATH_IMAGE006
Generating a pseudo rule base
Figure 244048DEST_PATH_IMAGE007
Said pseudo rule base
Figure 442948DEST_PATH_IMAGE007
The method is used for judging whether the enterprise to be declared has a tax cheating risk or not;
according to the pseudo rule base
Figure 846247DEST_PATH_IMAGE007
Generating a rule base
Figure 495403DEST_PATH_IMAGE008
And the rule base is used for scoring the fraud behaviors of the enterprise to be declared.
In some embodiments, the audit module is to base the rule base on
Figure 249733DEST_PATH_IMAGE008
Scoring the fraud behaviors of the enterprise to be declared;
according to the rule base
Figure 111378DEST_PATH_IMAGE086
And judging whether the project to be declared passes the audit or not according to the scoring result.
In some embodiments, the database module is configured to store case samples that include business data of a historical reporting enterprise, export data of a reporting project, and audit results of the reporting project.
In some embodiments, the case samples are derived from the result of discriminant publishing of the history declaration enterprise by a financial institution such as a bank in big data on one hand, and derived from the operation data of the enterprise to be declared and the export data of the project to be declared which are input into the system in real time on the other hand.
In some embodiments, case samples in the database module are updated periodically, the case samples maintain the effectiveness of the case samples in the database module for a certain period of time, and are periodically purged for expired case samples or non-native case samples.
In some embodiments, the self-learning module is configured to periodically instruct the data collation module and the rule generation module to calculate the pseudo rule base corresponding to the case sample according to the case sample stored in the database module
Figure 634764DEST_PATH_IMAGE087
And said rule base
Figure 64608DEST_PATH_IMAGE086
。
In some embodiments, the rule base module is for storing the pseudo rule base generated most recently according to the indication of the self-learning module
Figure 926076DEST_PATH_IMAGE087
And said rule base
Figure 466779DEST_PATH_IMAGE086
。
In some embodiments, the self-learning module continuously regenerates the pseudo rule base based on the latest version of the case sample
Figure 500463DEST_PATH_IMAGE087
And rule base
Figure 835629DEST_PATH_IMAGE086
In the original pseudo rule base
Figure 423605DEST_PATH_IMAGE087
And rule base
Figure 502420DEST_PATH_IMAGE086
On the basis of comparison, the rule base module is updated, so that the rule base always stores the newly generated pseudo rule base
Figure 390610DEST_PATH_IMAGE087
And rule base
Figure 896678DEST_PATH_IMAGE086
。
In a specific implementation, the present invention further provides a computer storage medium, where the computer storage medium may store a program, and the program may include some or all of the steps in the embodiments of the method and system for auditing export tax refunds provided by the present invention when executed. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a Random Access Memory (RAM).
Those skilled in the art will readily appreciate that the techniques of the embodiments of the present invention may be implemented as software plus a required general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be essentially or partially implemented in the form of a software product, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.
The foregoing description, for purposes of explanation, has been presented in conjunction with specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or to limit the embodiments to the precise forms disclosed above. Many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles and the practical application, to thereby enable others skilled in the art to best utilize the embodiments and various embodiments with various modifications as are suited to the particular use contemplated.

Claims (10)

1. An auditing method for export tax refunds is characterized by comprising the following steps:
acquiring operation data of an enterprise to be declared and export data of a project to be declared, wherein the project to be declared is a project which the enterprise to be declared expects to apply for export tax refund loan;
generating a transaction detail set according to the operation data of the enterprise to be declared and the export data of the project to be declared
Figure 18643DEST_PATH_IMAGE001
Said transaction detail set
Figure 60417DEST_PATH_IMAGE001
Including a number of transaction attributes of the enterprise to be declared
Figure 840154DEST_PATH_IMAGE002
Said transaction attribute
Figure 235363DEST_PATH_IMAGE002
Including a number of transaction items
Figure 177299DEST_PATH_IMAGE003
And with said transaction item
Figure 999761DEST_PATH_IMAGE003
Corresponding transaction parameters, wherein,s=1，2，3，……n，i=1，2，3，……n；
according to the transaction item
Figure 391429DEST_PATH_IMAGE003
And the transaction item
Figure 590329DEST_PATH_IMAGE003
Calculating transaction attributes corresponding to the transaction parameters
Figure 383841DEST_PATH_IMAGE002
Concentration ratio of
Figure 377205DEST_PATH_IMAGE004
；
According to the transaction attributes
Figure 131534DEST_PATH_IMAGE002
Concentration ratio of
Figure 993180DEST_PATH_IMAGE004
The transaction attribute is added
Figure 516565DEST_PATH_IMAGE002
Classifying to obtain common transaction item set
Figure 73973DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 315599DEST_PATH_IMAGE006
；
According to the common transaction item set
Figure 715356DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 358827DEST_PATH_IMAGE006
Generating a pseudo rule base
Figure 818627DEST_PATH_IMAGE007
Said pseudo rule base
Figure 547549DEST_PATH_IMAGE007
The enterprise tax cheating risk reporting system is used for judging whether the enterprise to be declared has tax cheating risks or not;
according to the pseudo rule base
Figure 750997DEST_PATH_IMAGE007
Generating a rule base
Figure 248974DEST_PATH_IMAGE008
The rule base is used for scoring the fraud behaviors of the enterprise to be declared;
according to the rule base
Figure 20621DEST_PATH_IMAGE008
Scoring the fraud behaviors of the enterprise to be declared;
according to the rule base
Figure 98823DEST_PATH_IMAGE008
And judging whether the project to be declared passes the audit or not according to the scoring result.
2. The method of claim 1, wherein the transaction is based on the transaction item
Figure 981328DEST_PATH_IMAGE003
And the transaction item
Figure 458446DEST_PATH_IMAGE003
Calculating transaction attributes corresponding to the transaction parameters
Figure 400994DEST_PATH_IMAGE002
Concentration ratio of
Figure 838929DEST_PATH_IMAGE004
The method comprises the following steps:
for each of said transaction attributes separately
Figure 384180DEST_PATH_IMAGE002
The transaction parameters in (1) are sequenced to obtain the transaction attributes
Figure 856749DEST_PATH_IMAGE002
An equal number of transaction parameter ordering sets;
according to the transaction item
Figure 94833DEST_PATH_IMAGE003
Sorting the transaction parameters according to the sequence from big to small according to the sizes of the corresponding transaction parameters to generate a transaction parameter sorting set;
according to
Figure 20063DEST_PATH_IMAGE009
Acquiring the concentration corresponding to the trading parameters in each trading parameter sequencing set
Figure 244371DEST_PATH_IMAGE004
Wherein
Figure 433432DEST_PATH_IMAGE010
is any one of the trade parameters in the sorted set of trade parameters,
Figure 717783DEST_PATH_IMAGE011
is the number of items corresponding to the trading parameters in the trading parameter ordering set,
Figure 254943DEST_PATH_IMAGE012
is that
Figure 282942DEST_PATH_IMAGE011
Is measured.
3. The method of claim 2, wherein said determining is based on said transaction attributes
Figure 589158DEST_PATH_IMAGE002
Concentration ratio of
Figure 778831DEST_PATH_IMAGE004
The transaction attribute is added
Figure 803288DEST_PATH_IMAGE002
Classifying to obtain common transaction item set
Figure 369399DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 671067DEST_PATH_IMAGE006
The method comprises the following steps:
obtaining the concentration degree in each transaction parameter sequencing set
Figure 159205DEST_PATH_IMAGE004
Maximum value of
Figure 546324DEST_PATH_IMAGE013
The transaction parameter of (a);
obtaining the concentration
Figure 40759DEST_PATH_IMAGE004
Maximum value of
Figure 931354DEST_PATH_IMAGE013
The number of items of the trading parameter arranged in the trading parameter order setx’；
For each of the transaction attributes
Figure 853043DEST_PATH_IMAGE002
The transaction item of
Figure 461879DEST_PATH_IMAGE003
Classifying the obtainedTransaction attributes
Figure 760005DEST_PATH_IMAGE002
Is located at the firstx’Sum of terms is less thanx’The transaction item corresponding to the item
Figure 770686DEST_PATH_IMAGE003
Is determined as a first set
Figure 738642DEST_PATH_IMAGE014
The transaction attribute is added
Figure 962338DEST_PATH_IMAGE002
Is located more than secondx’The transaction item corresponding to the item
Figure 673942DEST_PATH_IMAGE003
Is determined as a second set
Figure 663763DEST_PATH_IMAGE015
；
According to the first set
Figure 802621DEST_PATH_IMAGE014
And the second set
Figure 510683DEST_PATH_IMAGE015
Obtaining the common transaction item set
Figure 760398DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 745672DEST_PATH_IMAGE006
Wherein the common transaction item set
Figure 180064DEST_PATH_IMAGE005
Is the first setCombination of Chinese herbs
Figure 250789DEST_PATH_IMAGE014
The intersection of items in, the set of uncommon transaction items
Figure 431759DEST_PATH_IMAGE006
For the transaction detail set
Figure 271539DEST_PATH_IMAGE001
With the common transaction item set
Figure 752199DEST_PATH_IMAGE005
The difference of (a).
4. The method of claim 3, wherein the common set of transaction items is based on the common set of transaction items
Figure 169273DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 26371DEST_PATH_IMAGE006
Generating a pseudo rule base
Figure 845291DEST_PATH_IMAGE007
Said pseudo rule base
Figure 496853DEST_PATH_IMAGE007
The method is used for judging whether the enterprise to be declared has a tax fraud risk, and comprises the following steps:
according to the common transaction item set
Figure 542169DEST_PATH_IMAGE005
The transaction item of (1)
Figure 62012DEST_PATH_IMAGE003
Corresponding to the transaction parameters, obtaining the minimum support
Figure 610805DEST_PATH_IMAGE016
Wherein
Figure 802973DEST_PATH_IMAGE017
is the transaction item in the common transaction item set
Figure 70006DEST_PATH_IMAGE003
A corresponding minimum value of the transaction parameter;
according to the non-common transaction item set
Figure 268906DEST_PATH_IMAGE006
The transaction parameter corresponding to the transaction item in (1) obtains the minimum support degree
Figure 796839DEST_PATH_IMAGE018
Wherein
Figure 55782DEST_PATH_IMAGE019
is the set of uncommon transaction items
Figure 934746DEST_PATH_IMAGE006
The trade item corresponding to the median in (1)
Figure 671757DEST_PATH_IMAGE003
；
Setting confidence parameter valuesconfidenceAnd a value of a lower limit parameter of the regular lengthlenGenerating the common transaction item set according to Apriori algorithm
Figure 195143DEST_PATH_IMAGE005
Corresponding first set of rulers
Figure 749621DEST_PATH_IMAGE020
And the set of uncommon transaction items
Figure 725667DEST_PATH_IMAGE006
Corresponding second set of rules
Figure 393933DEST_PATH_IMAGE021
Wherein
Figure 37404DEST_PATH_IMAGE022
for a number of the first set of regulars,
Figure 497204DEST_PATH_IMAGE023
for a number of the second set of regulars,
Figure 226126DEST_PATH_IMAGE024
is (0, 1) type data when
Figure 304940DEST_PATH_IMAGE024
If not less than 0, judging that the enterprise to be declared has no tax fraud risk, and if so, judging that the enterprise to be declared has no tax fraud risk
Figure 927552DEST_PATH_IMAGE024
If =1, determining that the enterprise to be declared has a tax fraud risk;
according to the first rule body set
Figure 433619DEST_PATH_IMAGE025
And the second set of rules
Figure 774471DEST_PATH_IMAGE026
Generating a pseudo rule base
Figure 656976DEST_PATH_IMAGE027
。
5. The method of claim 4, wherein said determining is based on said pseudo rule base
Figure 137024DEST_PATH_IMAGE007
Generating a rule base
Figure 79572DEST_PATH_IMAGE008
The method comprises the following steps:
judging whether the enterprise to be declared is a historical tax cheating enterprise or not according to whether the enterprise to be declared has the historical tax cheating behavior or not;
according to the transaction parameters and the pseudo rule base
Figure 517506DEST_PATH_IMAGE007
Each rule body judges whether the enterprise to be declared has a tax fraud risk;
if the target rule body judges that the enterprise to be declared has no tax fraud risk, namely
Figure 62757DEST_PATH_IMAGE028
Then, set the initial value
Figure 535327DEST_PATH_IMAGE029
And calculate
Figure 507831DEST_PATH_IMAGE030
A value of (1), wherein
Figure 433062DEST_PATH_IMAGE031
Is the similarity of the target rule body and the transaction parameter, the target rule body being the pseudo rule base
Figure 782003DEST_PATH_IMAGE007
One of the rulers;
if the target rule body judges that the enterprise to be declared has the tax fraud risk, namely
Figure 109080DEST_PATH_IMAGE032
When it is set to be initialValue of
Figure 393430DEST_PATH_IMAGE033
And calculate
Figure 933521DEST_PATH_IMAGE034
A value of (d);
s11: if the enterprise to be declared does not have historical tax cheating behaviors, judging that the enterprise to be declared is not the historical tax cheating enterprise;
each one obtained by calculation
Figure 695940DEST_PATH_IMAGE030
Value of (A) and
Figure 2157DEST_PATH_IMAGE034
is brought into
Figure 457409DEST_PATH_IMAGE035
To calculate
Figure 481865DEST_PATH_IMAGE036
And is given a value of
Figure 47976DEST_PATH_IMAGE037
=
Figure 84065DEST_PATH_IMAGE038
Wherein
Figure 834852DEST_PATH_IMAGE039
when the enterprise to be declared is not the historical fraud enterprise, the fraud score of the enterprise to be declared,
Figure 221971DEST_PATH_IMAGE036
in order for the parameters to be updated,ta preset fraud score threshold;
according to calculation
Figure 453757DEST_PATH_IMAGE040
Value of (2), update
Figure 609932DEST_PATH_IMAGE037
；
Will be updated
Figure 531620DEST_PATH_IMAGE041
In (1)
Figure 140456DEST_PATH_IMAGE024
To be treated and
Figure 313949DEST_PATH_IMAGE024
is replaced by
Figure 449264DEST_PATH_IMAGE042
Wherein
Figure 417220DEST_PATH_IMAGE042
representing fraud scores of the to-be-declared enterprises;
s12: if the enterprise to be declared has historical tax cheating behaviors, judging that the enterprise to be declared is the historical tax cheating enterprise;
each one obtained by calculation
Figure 637986DEST_PATH_IMAGE030
Value of (A) and
Figure 84010DEST_PATH_IMAGE034
is brought into
Figure 214777DEST_PATH_IMAGE043
To calculate
Figure 481198DEST_PATH_IMAGE040
And is given a value of
Figure 64626DEST_PATH_IMAGE044
=
Figure 438976DEST_PATH_IMAGE045
Wherein
Figure 424249DEST_PATH_IMAGE046
when the enterprise to be declared is the historical fraud enterprise, the fraud score of the enterprise to be declared,
Figure 734008DEST_PATH_IMAGE036
in order for the parameters to be updated,tis a preset fraud score threshold value,
Figure 929366DEST_PATH_IMAGE042
representing fraud scores of the to-be-declared enterprises;
according to calculation
Figure 982773DEST_PATH_IMAGE040
Value of (2), update
Figure 947186DEST_PATH_IMAGE044
(ii) a Will be updated
Figure 427846DEST_PATH_IMAGE047
In (1)
Figure 847851DEST_PATH_IMAGE024
To be treated and
Figure 704948DEST_PATH_IMAGE024
is replaced by
Figure 399235DEST_PATH_IMAGE042
；
Recording
Figure 175430DEST_PATH_IMAGE042
And repeatedly performing S11 or S12, calculating the ratio of the times that the enterprise to be declared is performed S11 or S12gAnd, calculating
Figure 220746DEST_PATH_IMAGE007
In
Figure 740589DEST_PATH_IMAGE024
Is filled in
Figure 289382DEST_PATH_IMAGE037
Or
Figure 502058DEST_PATH_IMAGE044
The number of times of (2) is greater than or equal to 20h；
When in useg>=0.95 andh>=0.99, stop execution of S11 or S12, get rule base
Figure 769091DEST_PATH_IMAGE048
。
6. The method of claim 5, wherein the step of applying the coating is performed in a batch processg>=0.95 andh>=0.99, stop execution of S11 or S12, get rule base
Figure 702412DEST_PATH_IMAGE048
The method also comprises the following steps:
when in useg>=0.95 andh>=0.99, execution of S11 or S12 is stopped;
deleting
Figure 498854DEST_PATH_IMAGE007
Inlen（
Figure 757797DEST_PATH_IMAGE042
）<10 of
Figure 371181DEST_PATH_IMAGE007
；
Order to
Figure 373772DEST_PATH_IMAGE042
Obtained for the last 20 times
Figure 21791DEST_PATH_IMAGE042
To obtain a rule base
Figure 186056DEST_PATH_IMAGE048
。
7. The method of claim 6, wherein said rule base is based on
Figure 552316DEST_PATH_IMAGE008
And scoring the fraud behaviors of the enterprise to be declared, and further comprising the following steps:
according to the rule base
Figure 93019DEST_PATH_IMAGE008
Acquiring the information obtained when the enterprise to be declared is the historical tax deception enterprise
Figure 434411DEST_PATH_IMAGE042
Or obtaining the result obtained when the enterprise to be declared is not the historical tax deception enterprise
Figure 35157DEST_PATH_IMAGE042
，
Figure 498499DEST_PATH_IMAGE042
Scoring for fraud.
8. The method of claim 7, wherein said rule base is based on
Figure 967527DEST_PATH_IMAGE008
Scoring results ofJudging whether the project to be declared passes the audit, and further comprising:
if the fraud score is given
Figure 465504DEST_PATH_IMAGE049
If the enterprise to be declared is a tax cheating enterprise, the enterprise does not pass the audit;
if it is
Figure 96206DEST_PATH_IMAGE050
If the enterprise to be declared is a non-fraud enterprise, the audit is passed.
9. An audit system for an export tax returned loan, applied to the method of any of claims 1-8, the system comprising:
a data sorting module: the method comprises the steps of obtaining operation data of an enterprise to be declared and export data of a project to be declared, wherein the project to be declared is a project which the enterprise to be declared expects to apply for export tax refund loan;
generating a transaction detail set according to the operation data of the enterprise to be declared and the export data of the project to be declared
Figure 312423DEST_PATH_IMAGE001
Said transaction detail set
Figure 319562DEST_PATH_IMAGE001
Including a number of transaction attributes of the enterprise to be declared
Figure 672046DEST_PATH_IMAGE002
Said transaction attribute
Figure 349015DEST_PATH_IMAGE002
Including a number of transaction items
Figure 914513DEST_PATH_IMAGE003
And with said transaction item
Figure 600710DEST_PATH_IMAGE003
Corresponding transaction parameters, wherein,s=1，2，3，……n，i=1，2，3，……n；
a rule generation module: for use in accordance with the transaction item
Figure 932334DEST_PATH_IMAGE003
And the transaction item
Figure 45783DEST_PATH_IMAGE003
Calculating transaction attributes corresponding to the transaction parameters
Figure 95648DEST_PATH_IMAGE002
Concentration ratio of
Figure 319956DEST_PATH_IMAGE004
；
According to the transaction attributes
Figure 771666DEST_PATH_IMAGE002
Concentration of said transaction attributes
Figure 56017DEST_PATH_IMAGE002
Classifying to obtain common transaction item set
Figure 202964DEST_PATH_IMAGE005
And an uncommon transaction item set
Figure 358526DEST_PATH_IMAGE006
；
According to the common transaction item set
Figure 540109DEST_PATH_IMAGE005
And the set of uncommon transaction items
Figure 119995DEST_PATH_IMAGE006
Generating a pseudo rule base
Figure 19818DEST_PATH_IMAGE007
Said pseudo rule base
Figure 710562DEST_PATH_IMAGE007
The enterprise tax cheating risk reporting system is used for judging whether the enterprise to be declared has tax cheating risks or not;
according to the pseudo rule base
Figure 746651DEST_PATH_IMAGE007
Generating a rule base
Figure 372805DEST_PATH_IMAGE008
The rule base is used for scoring the fraud behaviors of the enterprise to be declared;
an auditing module: for use in accordance with the rule base
Figure 618978DEST_PATH_IMAGE008
Scoring the fraud behaviors of the enterprise to be declared;
according to the rule base
Figure 988780DEST_PATH_IMAGE008
And judging whether the project to be declared passes the audit or not according to the scoring result.
10. The system of claim 9, further comprising a database module, a self-learning module, and a rule base module,
the database module is used for storing case samples, and the case samples comprise operation data of historical reporting enterprises, export data of reporting projects and auditing results of the reporting projects;
the self-learning module is used for indicating the data sorting module and the case samples stored in the database module periodicallyA rule generation module calculates the pseudo rule base corresponding to the case sample
Figure 272518DEST_PATH_IMAGE007
And said rule base
Figure 69573DEST_PATH_IMAGE008
；
The rule base module is used for storing the pseudo rule base which is generated latest according to the indication of the self-learning module
Figure 803042DEST_PATH_IMAGE007
And said rule base
Figure 710956DEST_PATH_IMAGE008
。
CN202210062850.9A 2022-01-20 2022-01-20 Auditing method and system for export tax refund loan Active CN114092215B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210062850.9A CN114092215B (en) 2022-01-20 2022-01-20 Auditing method and system for export tax refund loan

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210062850.9A CN114092215B (en) 2022-01-20 2022-01-20 Auditing method and system for export tax refund loan

Publications (2)

Publication Number Publication Date
CN114092215A true CN114092215A (en) 2022-02-25
CN114092215B CN114092215B (en) 2022-04-19

Family

ID=80308670

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210062850.9A Active CN114092215B (en) 2022-01-20 2022-01-20 Auditing method and system for export tax refund loan

Country Status (1)

Country Link
CN (1) CN114092215B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115018609A (en) * 2022-08-09 2022-09-06 南京币鑫数据科技有限公司 Export tax return declaration automation model and algorithm based on big data

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108230126A (en) * 2016-12-13 2018-06-29 上海海万信息科技股份有限公司 Taxation declaration, audit and management system based on O2O patterns
CN108596750A (en) * 2018-05-09 2018-09-28 西安交通大学 A kind of invoice voiding recognition methods based on parallel association rules
CN109345273A (en) * 2019-01-04 2019-02-15 南京擎天科技有限公司 A kind of neutralizing of export enterprise's financing information is determined method and system
CN110222957A (en) * 2019-05-20 2019-09-10 深圳壹账通智能科技有限公司 A kind of data checking method and relevant device
CN112613961A (en) * 2020-12-15 2021-04-06 航天信息股份有限公司海外业务分公司 Import and export enterprise tax collection and management system based on block chain

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108230126A (en) * 2016-12-13 2018-06-29 上海海万信息科技股份有限公司 Taxation declaration, audit and management system based on O2O patterns
CN108596750A (en) * 2018-05-09 2018-09-28 西安交通大学 A kind of invoice voiding recognition methods based on parallel association rules
CN109345273A (en) * 2019-01-04 2019-02-15 南京擎天科技有限公司 A kind of neutralizing of export enterprise's financing information is determined method and system
CN110222957A (en) * 2019-05-20 2019-09-10 深圳壹账通智能科技有限公司 A kind of data checking method and relevant device
CN112613961A (en) * 2020-12-15 2021-04-06 航天信息股份有限公司海外业务分公司 Import and export enterprise tax collection and management system based on block chain

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115018609A (en) * 2022-08-09 2022-09-06 南京币鑫数据科技有限公司 Export tax return declaration automation model and algorithm based on big data
CN115018609B (en) * 2022-08-09 2022-11-22 南京币鑫数据科技有限公司 Export tax return declaration automation system and method based on big data

Also Published As

Publication number Publication date
CN114092215B (en) 2022-04-19

Similar Documents

Publication Publication Date Title
Engelmann et al. The Basel II risk parameters: estimation, validation, and stress testing
Finlay Credit scoring, response modeling, and insurance rating: a practical guide to forecasting consumer behavior
US7296734B2 (en) Systems and methods for scoring bank customers direct deposit account transaction activity to match financial behavior to specific acquisition, performance and risk events defined by the bank using a decision tree and stochastic process
Doumpos et al. Analytical techniques in the assessment of credit risk
Dimitras et al. Evaluation of empirical attributes for credit risk forecasting from numerical data
CN114092215B (en) Auditing method and system for export tax refund loan
Donovan et al. Determining credit risk using qualitative disclosure
Wang et al. Improving investment suggestions for peer-to-peer lending via integrating credit scoring into profit scoring
CN112085593B (en) Credit data mining method for small and medium enterprises
US20220398583A1 (en) Transaction reconciliation and deduplication
CN115860924A (en) Supply chain financial credit risk early warning method and related equipment
Katsimperis et al. Creating a flexible business credit rating model using multicriteria decision analysis
Pradnyana et al. Loan Default Prediction in Microfinance Group Lending with Machine Learning
Bhagat Predicting Loan Defaults using Machine Learning Techniques
Rahmat et al. Prediction of Regional Revenue and Expenditure Budget using Autoregressive Integrated Moving Average (ARIMA)
Alaba et al. Data mining algorithm for development of a predictive model for mitigating loan risk in Nigerian banks
Melnyk et al. Improving the quality of credit activity by using scoring model
Verster et al. Predicting take-up of home loan offers using tree-based ensemble models: A South African case study
US20230419404A1 (en) Methods and Systems for Determining Segmentation Rules for Categorising Entities
CN117291740B (en) Receivables data authenticity intelligent identification auditing system based on big data
Roijmans Macroeconomic factors in loan default prediction
Şakar Variable Importance Analysis in Default Prediction using Machine Learning Techniques
CN117994017A (en) Method for constructing retail credit risk prediction model and online credit service Scoredelta model
PRODUCT et al. ISTANBUL TECHNICAL UNIVERSITY★ GRADUATE SCHOOL
Danenas et al. Intelligent techniques and systems in credit risk analysis and forecasting: a review of patents

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
GR01 Patent grant
GR01 Patent grant