CN109635007A - A kind of behavior evaluation method, apparatus and relevant device - Google Patents
A kind of behavior evaluation method, apparatus and relevant device Download PDFInfo
- Publication number
- CN109635007A CN109635007A CN201811550706.XA CN201811550706A CN109635007A CN 109635007 A CN109635007 A CN 109635007A CN 201811550706 A CN201811550706 A CN 201811550706A CN 109635007 A CN109635007 A CN 109635007A
- Authority
- CN
- China
- Prior art keywords
- information
- enterprise
- initial
- personnel
- behavior
- 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
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 claims abstract description 45
- 238000012216 screening Methods 0.000 claims abstract description 29
- 238000000605 extraction Methods 0.000 claims abstract description 10
- 230000006399 behavior Effects 0.000 claims description 118
- 208000015181 infectious disease Diseases 0.000 claims description 44
- 238000004364 calculation method Methods 0.000 claims description 25
- 238000012795 verification Methods 0.000 claims description 21
- 238000004422 calculation algorithm Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 6
- 238000010845 search algorithm Methods 0.000 claims description 6
- 230000003542 behavioural effect Effects 0.000 claims 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000007726 management method Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 238000012854 evaluation process Methods 0.000 description 5
- 238000011144 upstream manufacturing Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000010354 integration Effects 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 238000011835 investigation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application discloses a kind of behavior evaluation methods, and the target data information including the Target Enterprise to known behavior type carries out information extraction, obtain initial association company information and initial association personal information；Infectiosity calculating is carried out to initial association enterprise, cohesion calculating is carried out to initial association personnel, to obtain intermediate affiliated enterprise's information and intermediate associate people information；According to intermediate affiliated enterprise's information and the big figure of intermediate associate people information architecture related network, and it is associated degree screening to it, obtains affiliated enterprise and associate people, affiliated enterprise and associate people are labeled as the corresponding behavior type of Target Enterprise；This method can quickly and accurately excavate entire association group according to the relevant information of the enterprise of specific behavior type, to realize the effective assessment and unified management of entire group, enterprise.Disclosed herein as well is a kind of behavior evaluation device, equipment and computer readable storage mediums, all have above-mentioned beneficial effect.
Description
Technical Field
The present application relates to the field of internet technologies, and in particular, to a behavior evaluation method, and further, to a behavior evaluation device, a behavior evaluation apparatus, and a computer-readable storage medium.
Background
Generally, the operation of an enterprise is less than the cooperation of multiple enterprises and the participation of multiple personnel, so how to quickly and accurately find the enterprise or the personnel having a certain association relation with the enterprise based on a certain specific enterprise to realize the comprehensive evaluation of the whole enterprise operation group and the unified management and supervision is always a more complex problem.
In the prior art, all related other enterprises and persons are mostly obtained by collecting related enterprise information of a specific enterprise and inquiring, and then professional departments identify and evaluate individuals one by one, so that an enterprise group with a large association relation is identified and obtained, and the management of the whole enterprise operation group is realized. However, this implementation is very labor intensive and inefficient. More specifically, for some enterprises, in order to meet certain requirements of the enterprises, a false means is adopted in the operation process, and the conditions of violating law and regulation behaviors, such as tax evasion, tax omission, invoice false opening and the like, can only supervise and check related enterprises after the illegal behaviors occur, but cannot effectively find the illegal behaviors in advance by means of original data, and the single-point breakthrough mode causes that the checked and checked illegal enterprises are fewer and the illegal behaviors are frequently escaped when the checked and controlled. Therefore, the existing enterprise behavior assessment can only perform classification identification and assessment on individuals, but cannot trace back the behavior information of the whole group, and is low in efficiency and accuracy.
Therefore, how to rapidly and accurately excavate the whole association group according to the related information of the enterprise with a specific behavior type, so as to achieve effective evaluation and unified management of the whole enterprise group is a problem to be solved by those skilled in the art.
Disclosure of Invention
The method can rapidly and accurately excavate the whole association group according to the related information of enterprises with specific behavior types, thereby realizing effective evaluation and unified management of the whole enterprise group; another object of the present application is to provide a behavior evaluation device, an apparatus, and a computer-readable storage medium, which also have the above-mentioned advantageous effects.
In order to solve the above technical problem, the present application provides a behavior evaluation method, including:
acquiring target data information of a target enterprise; wherein the target enterprise is an enterprise of a known behavior type;
extracting information from the target data information to obtain initial associated enterprise information and initial associated personnel information;
calculating the infection degree of the initial associated enterprise according to the initial associated enterprise information, and extracting intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information;
performing intimacy calculation on initial associated personnel according to the initial associated personnel information, and extracting intermediate associated personnel information meeting a preset intimacy threshold value condition from the initial associated personnel information;
constructing a correlation network big graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information;
screening the association degree of the association network large graph to obtain an association enterprise and an association person;
and marking the associated enterprises and the associated personnel as behavior types corresponding to the target enterprises.
Preferably, the extracting information of the target data information to obtain initial associated enterprise information and initial associated personnel information includes:
acquiring all related enterprise information and all related personnel information according to the target data information;
integrating all the related enterprise information and the related personnel information to generate a two-dimensional broad table;
importing the two-dimensional wide table into an ArangoDB graph database for processing to obtain an initial association network;
and extracting the initial associated enterprise information and the initial associated personnel information in a preset range of the initial associated network.
Preferably, the integrating all the related enterprise information and the related personnel information to generate the two-dimensional wide table includes:
constructing a triple transaction edge of a target enterprise-relation-related enterprise according to all the related enterprise information;
constructing a first triple control edge of the related personnel-relationship-target enterprise according to all the related personnel information;
constructing a second triple control edge of the related personnel-relationship-related enterprise according to the triple transaction edge and the first triple control edge;
and integrating the triple transaction edge, the first triple control edge and the second triple control edge to obtain the two-dimensional wide table.
Preferably, the extracting the initial associated enterprise information and the initial associated person information within the preset range of the initial associated network includes:
and extracting the initial associated enterprise information and the initial associated personnel information in a preset range of the initial associated network by using a breadth-first search algorithm.
Preferably, the screening of the association degree of the association network large graph to obtain the association enterprises and the association personnel includes:
calculating the associated network big graph by using a connected community division algorithm to obtain a connected subgraph set;
screening all connected subgraphs in the connected subgraph set through a preset service rule to obtain associated subgraphs;
and extracting the associated enterprises and the associated persons from the associated subgraph.
Preferably, the fraudulent group identification method further comprises:
performing service verification on the associated subgraph to obtain a verification result;
and determining a group operation mode according to the verification result by combining a preset connectivity mode.
Preferably, the fraudulent group identification method further comprises:
feeding back and adjusting parameters according to the verification result; the adjustment parameters comprise the preset infection threshold, the preset intimacy threshold, the preset range of the initial association network, the preset business rule and the preset communication relation mode.
In order to solve the above technical problem, the present application further provides a behavior evaluation device, including:
the information acquisition module is used for acquiring target data information of a target enterprise; wherein the target enterprise is an enterprise of a known behavior type;
the information extraction module is used for extracting information from the target data information to obtain initial associated enterprise information and initial associated personnel information;
the infection degree calculation module is used for calculating the infection degree of the initial associated enterprise according to the initial associated enterprise information and extracting intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information;
the intimacy degree calculation module is used for carrying out intimacy degree calculation on initial associated personnel according to the initial associated personnel information and extracting intermediate associated personnel information meeting a preset intimacy degree threshold value condition from the initial associated personnel information;
the network construction module is used for constructing a correlation network big graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information;
the network screening module is used for screening the association degree of the associated network large graph to obtain associated enterprises and associated personnel;
and the behavior marking module is used for marking the associated enterprises and the associated personnel as behavior types corresponding to the target enterprises.
In order to solve the above technical problem, the present application further provides a behavior evaluation device, including:
a memory for storing a computer program;
a processor for implementing the steps of any of the above behavior assessment methods when executing the computer program.
In order to solve the above technical problem, the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any one of the behavior assessment methods.
The behavior evaluation method comprises the steps of obtaining target data information of a target enterprise; wherein the target enterprise is an enterprise of a known behavior type; extracting information from the target data information to obtain initial associated enterprise information and initial associated personnel information; calculating the infection degree of the initial associated enterprise according to the initial associated enterprise information, and extracting intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information; performing intimacy calculation on initial associated personnel according to the initial associated personnel information, and extracting intermediate associated personnel information meeting a preset intimacy threshold value condition from the initial associated personnel information; constructing a correlation network big graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information; screening the association degree of the association network large graph to obtain an association enterprise and an association person; and marking the associated enterprises and the associated personnel as behavior types corresponding to the target enterprises.
Therefore, according to the behavior evaluation method provided by the application, according to the related data information of the target enterprise with the specific behavior type, the infectivity and the affinity of all related enterprises and persons are calculated, so that the construction of the associated network big graph is completed, further, the associated network big graph is subjected to association degree screening, other enterprises and persons with the same behavior type as the target enterprise can be identified and obtained, and therefore, the enterprises and persons with the same period and relevance can be mined out based on the enterprise with the specific behavior type of one user, the identification of the group with the same behavior type is realized, the group can be effectively evaluated and uniformly managed by professional departments, and the real-time performance and the pertinence are high; in addition, the enterprise behavior evaluation is realized based on the computer technology, and compared with the method for carrying out classification evaluation on individuals through manual investigation in the prior art, the method has higher evaluation efficiency and evaluation accuracy.
The behavior evaluation device, the equipment and the computer-readable storage medium provided by the application all have the beneficial effects, and are not described herein again.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic flow chart of a behavior evaluation method provided in the present application;
fig. 2 is a schematic flow chart of a network large graph screening method in a behavior evaluation process according to the present application;
fig. 3 is a schematic flow chart of a method for extracting relevant information in a behavior evaluation process according to the present application;
fig. 4 is a schematic structural diagram of a behavior evaluation device provided in the present application;
fig. 5 is a schematic structural diagram of a behavior evaluation device provided in the present application.
Detailed Description
The core of the application is to provide a behavior evaluation method, which can rapidly and accurately excavate the whole association group according to the related information of enterprises with known behavior types, thereby realizing the effective evaluation and unified management of the whole group; another core of the present application is to provide a behavior assessment apparatus, a device and a computer-readable storage medium, which also have the above-mentioned advantages.
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Referring to fig. 1, fig. 1 is a schematic flow chart of a behavior evaluation method provided in the present application, where the behavior evaluation method may include:
s101: acquiring target data information of a target enterprise; wherein the target enterprise is an enterprise with a known behavior type;
specifically, the identification and evaluation of the enterprise associated group with a specific behavior type mainly depends on a certain enterprise determined to have the specific behavior, and further data mining and searching are performed on the basis of the enterprise, so that the corresponding enterprise associated group can be obtained, and therefore group management and supervision are achieved. Therefore, when identifying and evaluating an enterprise-related group having a certain specific behavior, it is first necessary to acquire related data of an enterprise for which the specific behavior is determined to exist, that is, the target enterprise, and of course, the related data is data having a certain association relationship with the specific behavior, that is, the target data information. The type of the target data information corresponds to the behavior type of the target enterprise, for example, for an enterprise with tax evasion behavior, the target data information is the tax payment information of the enterprise, and for an enterprise with false invoice behavior, the target data information is the invoice information of the enterprise. The above method for acquiring the target data information may be implemented based on any one of the existing technologies, which is not limited in this application.
S102: extracting information of the target data information to obtain initial associated enterprise information and initial associated personnel information;
the step aims to extract information of the target data information when the target data information is obtained so as to obtain information of other enterprises and persons having a certain incidence relation with the target enterprise, namely the initial associated enterprise information and the initial associated person information. For example, when a target enterprise has a false invoice issuing behavior, the invoice information is target data information, which includes information such as transaction data, information of supply and marketing parties, financial responsible persons, ticket purchasing persons, and the like, so that all relevant enterprises and persons suspected to participate in the false invoice issuing behavior of the target enterprise, including upstream relevant enterprises and persons, downstream relevant enterprises and persons of the target enterprise, can be extracted from the target data information; further, the enterprise and the personnel which have a relationship with the target enterprise virtual invoice making behavior to a certain extent can be extracted from all the related enterprises and personnel, so as to obtain the corresponding initial related enterprise information and initial related personnel information, for example, the related enterprises at the upstream layer and the downstream layer of the target enterprise can be extracted as the initial related enterprises, and the related personnel at the downstream layer of the target enterprise can be extracted as the initial related personnel.
S103: calculating the infection degree of the initial associated enterprise according to the initial associated enterprise information, and extracting intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information;
after the initial associated enterprise information is obtained, the infection degree calculation is carried out on the corresponding initial associated enterprises based on the initial associated enterprise information, so that intermediate associated enterprises with higher association degree with target enterprises are extracted from all the initial associated enterprises, and corresponding related information, namely the intermediate associated enterprise information, is obtained; wherein the intermediate associated enterprise has a higher degree of association with the target enterprise than the initial associated enterprise. Specifically, the higher the infection degree is, the higher the association degree between the initial associated enterprise and the target enterprise is, so that the infection degrees of all the initial associated enterprises compared with the target enterprise can be calculated, and the initial associated enterprise with the infection degree higher than the preset infection degree threshold value is extracted as the intermediate associated enterprise, thereby further obtaining the corresponding intermediate associated enterprise information. Of course, the specific value of the preset infection threshold may be set by a technician according to actual needs, which is not limited in this application.
S104: performing intimacy calculation on the initial associated personnel according to the initial associated personnel information, and extracting intermediate associated personnel information meeting a preset intimacy threshold value condition from the initial associated personnel information;
after the initial associated person information is obtained, performing affinity calculation on corresponding initial associated persons based on the initial person information, so as to extract and obtain intermediate associated persons with higher association degree with the target enterprise from all the initial associated persons, and obtain corresponding related information, namely the intermediate associated person information; wherein the degree of association between the intermediate associated person and the target business is higher than the initial associated person. Specifically, the higher the intimacy degree is, the higher the degree of association between the initial associated person and the target enterprise is, so that the intimacy degree of all the initial associated persons compared with the target enterprise can be calculated, and the initial associated person with the intimacy degree higher than a preset intimacy degree threshold value is extracted as the intermediate associated person, so that the corresponding intermediate associated person information is further obtained. Similarly, the specific value of the preset intimacy degree threshold value may be set by a technician according to actual needs, which is not limited in the present application.
It should be noted that, S103 and S104 respectively implement the infection degree calculation for the initial related enterprise and the intimacy calculation for the initial related personnel, but the implementation order of the above two steps does not affect the implementation of the present technical solution, and certainly, to ensure the evaluation efficiency, the two steps can be executed simultaneously.
S105: constructing a correlation network big graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information;
specifically, the step aims to construct an association network large graph with a higher association degree with the target enterprise based on the intermediate association enterprise information and the intermediate association personnel information, so that other enterprises or personnel with higher similarity to the target enterprise behavior type can be obtained according to more accurate identification of the association network large graph in the subsequent process, and the identification of the enterprise association group is completed. The relational network large graph can clearly represent the relation information among the target enterprise, the intermediate relational enterprise and the intermediate relational personnel, and the identification and evaluation of the enterprise relational group are more convenient. Of course, any one of the prior arts can be adopted for the method for constructing the associated network large graph, and the application is not limited.
S106: screening the association degree of the association network large graph to obtain association enterprises and association personnel;
s107: and marking the associated enterprises and the associated personnel as behavior types corresponding to the target enterprises.
Specifically, the steps are to further screen the associated network large graph to accurately obtain the associated enterprises and associated persons of the same behavior type as the target enterprise, that is, the associated enterprises are other identified enterprises participating in the specific behavior of the target enterprise, and the associated persons are identified related persons participating in the specific behavior of the target enterprise; further, the two can be marked as the corresponding behavior types of the target enterprises, thereby completing the identification and evaluation of the enterprise association groups with a certain specific behavior. In addition, the screening mode of the association degree can be realized based on a relevant algorithm or a business rule.
As a preferred embodiment, please refer to fig. 2, fig. 2 is a schematic flowchart of a network large graph screening method in a behavior evaluation process provided by the present application, where the screening of the association degree of the associated network large graph to obtain the associated enterprise and the associated personnel may include:
s201: calculating a big associated network graph by using a connected community division algorithm to obtain a connected subgraph set;
s202: screening all connected subgraphs in the connected subgraph set through a preset service rule to obtain associated subgraphs;
s203: and extracting the associated enterprises and the associated persons from the associated subgraph.
The application provides a more specific associated subgraph screening method, which comprises the steps of firstly carrying out calculation processing on an associated network big graph through a connected community division algorithm to obtain a connected subgraph set, wherein the connected subgraph set comprises a plurality of connected subgraphs with a certain association relation, the connected community division algorithm is an algorithm which is provided for reasonably dividing a complex network and finding out a real existing community structure, and the algorithm has higher division accuracy. Further, all connected subgraphs in the connected subgraph set can be filtered based on preset business rules, so that associated subgraphs are obtained, all enterprises and persons involved in the associated subgraphs correspond to the associated enterprises and the associated persons, and therefore the associated enterprises and the associated persons can be extracted and obtained from the associated subgraphs. Of course, the preset business rules can be set differently according to actual requirements, and mainly aim at the number of enterprise nodes and the degree of output of personnel nodes in the associated subgraph.
As a preferred embodiment, the behavior evaluation method may further include performing service verification on the associated subgraph to obtain a verification result; and determining a group operation mode according to a verification result by combining the preset connectivity mode.
Specifically, the embodiment aims to verify the identification result, that is, the associated subgraph, so as to obtain a corresponding verification result, and further, determines the group operation mode of the identified enterprise associated group in combination with a preset connectivity mode, so as to facilitate subsequent auditing work of relevant departments. The group operation mode can be divided into a transaction control triangle, a quadrilateral ring, a pentagonal ring and the like, the preset communication relation mode is the same as the above, and the specific type of the preset communication relation mode can be set according to actual business requirements.
The transaction control triangle is composed of a transaction relation side and two control relation sides; rings larger than the triangle may be closed rings of groups (such as fund flow), or closed rings of enterprises (such as transactions between a same enterprise and different related enterprises), and the relationship information in such rings is complex, and cannot be directly identified as a group partner, and further pattern recognition (such as occupation ratio of the number of people in the rings and the transaction relationship control relationship) needs to be performed depending on business rules. Furthermore, for the case where two persons control multiple same businesses (including more than 2 businesses with specific behaviors), the two persons may be considered to be one person (e.g., family relatives), and for such cases, the triangular ring of money return may be expanded to a quadrilateral, pentagonal ring, etc.
According to the behavior evaluation method provided by the application, according to the related data information of the target enterprise with the specific behavior type, the infectivity and the affinity of all related enterprises and persons are calculated, so that the construction of the associated network big graph is completed, further, the associated network big graph is subjected to association degree screening, and other enterprises and persons with the same behavior type as the target enterprise can be identified and obtained from the associated network big graph, so that the enterprises and persons with the same association can be mined out based on the enterprise with the specific behavior type of one user, the identification of the group with the same behavior type is realized, the effective evaluation and unified management of the group by professional departments are facilitated, and the real-time performance and the pertinence are higher; in addition, the enterprise behavior evaluation is realized based on the computer technology, and compared with the method for carrying out classification evaluation on individuals through manual investigation in the prior art, the method has higher evaluation efficiency and evaluation accuracy.
On the basis of the above embodiments, please refer to fig. 3, and fig. 3 is a schematic flow chart of a method for extracting relevant information in a behavior evaluation process according to the present application.
As a preferred embodiment, the process of extracting information from the target data information in S102 to obtain the initial associated enterprise information and the initial associated person information may specifically include:
s301: acquiring all related enterprise information and all related personnel information according to the target data information;
s302: integrating all related enterprise information and related personnel information to generate a two-dimensional wide table;
s303: importing the two-dimensional wide table into an ArangoDB graph database for processing to obtain an initial association network;
s304: and extracting initial associated enterprise information and initial associated personnel information in a preset range of the initial associated network.
The method for extracting the initial associated enterprise and the initial associated personnel is a more specific implementation mode. Specifically, all other enterprises and persons having an association relationship with the target enterprise, that is, all enterprises and persons involved in the target data information, may be extracted from the target data information, and then integrated to generate a two-dimensional wide table, so that the two-dimensional wide table covers all other enterprises and persons related to the target enterprise and their corresponding relationships; further, the two-dimensional broad table is imported into an ArangoDB graph database for processing, the ArangoDB supports a flexible data model and has high performance, the two-dimensional broad table is processed by the ArangoDB to obtain a corresponding initial association network, the initial association network is similar to the association network large graph, but the difference is that the relationship information among a target enterprise, an initial association enterprise and initial association personnel is clearly represented in the initial association network, and the relationship information among the target enterprise, intermediate association enterprises and intermediate association personnel is clearly represented in the association network large graph. Therefore, the initial associated enterprise and the initial associated personnel can be extracted within the preset range in the initial associated network. Of course, the setting of the preset range does not affect the implementation of the present technical solution, and referring to the example of the previous embodiment, the preset range may be related enterprises and people in the upstream layer and the downstream layer from the target enterprise.
Preferably, the integrating all the related enterprise information and the related personnel information to generate the two-dimensional wide table may include: constructing a triple transaction edge of the target enterprise-relation-related enterprise according to all related enterprise information; constructing a first triple control edge of the related personnel-relationship-target enterprise according to all related personnel information; constructing a second triple control edge of the related personnel-relationship-related enterprise according to the triple transaction edge and the first triple control edge; and integrating the triple transaction edge, the first triple control edge and the second triple control edge to obtain the two-dimensional width table.
For the acquisition process of the two-dimensional wide table in S302, the present application provides a more specific implementation manner. Specifically, the triple transaction edge of the target enterprise-relation-related enterprise and the triple control edge of the related personnel-relation-target enterprise can be constructed based on all related enterprise information and related personnel information, and the triple control edge of the related personnel-relation-related enterprise is further obtained according to the triple transaction edge and the related personnel-relation-related enterprise, so that the corresponding two-dimensional broad table can be obtained based on the integration of the triple transaction edge and the two triple control edges.
Preferably, the extracting the initial associated enterprise information and the initial associated person information within the preset range of the initial associated network may include extracting the initial associated enterprise information and the initial associated person information within the preset range of the initial associated network by using a breadth-first search algorithm.
Specifically, the method for extracting the initial associated enterprise information and the initial associated person information in S303 may be implemented based on a Breadth-First Search algorithm (BFS), where the BFS algorithm is a graph Search algorithm, and all nodes in a graph may be systematically expanded and checked to find a result, and the BFS algorithm has completeness, that is, regardless of the type of a graph, as long as a target exists, a target node, that is, an enterprise and a person with fraudulent behavior may be found.
Therefore, based on the process, the extraction of the initial associated enterprise information and the initial associated personnel information is realized, and the identification efficiency and the identification accuracy are higher.
On the basis of the foregoing embodiments, to further improve the recognition accuracy, the behavior evaluation method may further include: feeding back and adjusting parameters according to the verification result; the adjusting parameters comprise a preset infection threshold, a preset intimacy threshold, a preset range of an initial association network, a preset service rule and a preset communication relation mode.
The embodiment of the application aims to realize feedback adjustment of related parameters so as to further improve the accuracy of the identification algorithm. Specifically, the verification of the relevant parameters, that is, the preset parameters involved in the behavior evaluation process, may be implemented according to the verification result, and specifically may include a preset infection threshold, a preset intimacy threshold, a preset range of the initial association network, a preset business rule, a preset connectivity mode, and the like.
For other implementation steps in the embodiments of the present application, reference may be made to the previous embodiment, and details of the present application are not described herein again.
The behavior evaluation method provided by the embodiment of the application effectively realizes the adjustment of each preset relevant parameter in the identification process based on the identification result, improves the accuracy of the behavior identification algorithm, and further improves the accuracy of the corresponding evaluation result.
On the basis of the above embodiments, taking a target enterprise with a false invoice issuing behavior as an example, the application provides a more specific behavior evaluation method. The virtual enterprise in the following contents is the target enterprise.
(1) Processing original data (namely the target data information) of the invoice virtual-open enterprise, integrating and extracting data information including the virtual-open enterprise, the virtual-open invoice associated enterprise and personnel thereof, and acquiring related total data (namely the initial related enterprise information and the initial related personnel information):
the invoice virtual-open enterprise is a target enterprise which is identified and determined to have the invoice virtual-open behavior, and the original data of the invoice virtual-open enterprise is further input as an invoice virtual-open group identification algorithm for data processing.
In the embodiment, the pseudo-open enterprise is taken as a starting point, one layer is found at the upstream of the pseudo-open enterprise, three layers are found at the downstream, and all the obtained data of the related enterprises form triple transaction edges of the pseudo-open enterprise-relation-related enterprises.
The data information of the associated personnel can be obtained from data such as tax register data of taxpayers, investment relation data, natural person information and the like so as to obtain basic information, such as identity numbers, mobile phone numbers, attributions and the like of personnel associated with the invoice virtual opening enterprise, such as enterprise legal persons, investors, financial responsible persons, tax clerks, ticket buyers and the like, and therefore all the obtained associated personnel data form a triple control edge of the associated personnel-relation-virtual opening enterprise.
Further, according to the triple transaction edge and the triple control edge in the two steps, association between the associated personnel and the associated enterprise is carried out, and a triple control edge of the associated personnel-relationship-associated enterprise is formed. Therefore, the triple transaction side data and the two triple control side data can be integrated, and the entity data in the triple transaction side data and the two triple control side data are extracted to form a two-dimensional wide table.
Further, the two-dimensional broad table is imported into an ArangoDB database to form an initial association network of enterprises and personnel and enterprises. Specifically, firstly, integrating a triple transaction EDGE and a triple control EDGE into an EDGE EDGE table; secondly, extracting an enterprise in the triple transaction edge as an entity 1, extracting a person in the triple control edge as an entity 2 and an enterprise as an entity 3, and making an entity 1/2/3 into a node VERTEX point table; and finally, taking the EDGE EDGE table as the relation input and the VERTEX point table as the entity input to construct an initial association network of TH (group partner). Wherein, the relevant attributes of the VERTEX point table and the EDGE EDGE table can be set as follows:
VERTEX Point Table:
key: the enterprise node is the taxpayer electronic file number, and the personnel node is the identity card number;
sfxk: whether a false opening behavior exists, 1 is selected, and 0 is not selected; wherein, because the personnel are all related personnel of the virtual enterprise, the personnel nodes all take 1;
and (3) ishuman: distinguishing whether the nodes are personnel nodes or not, wherein the personnel nodes are 1, and the enterprise nodes are 0;
sensi _ weight: the infection degree is set to 0 in the initial value;
int _ weight: setting the initial values of the intimacy degree to be 0;
is _ community: whether the initial value is a virtual open group is set to be 0;
community _ label: the virtual open burst indicates that the initial values are all set to 0.
EDGE EDGE table:
from: the triple transaction side is the electronic document number of the seller taxpayer, and the triple control side is the identity card number of the responsible person;
a _ to: downstream enterprise taxpayer electronic file number;
je: supply and sale monthly information amount;
xfxkze: the total amount of the false invoices are made by the false-invoicing enterprises;
gfze: the total amount of the false invoice purchased by the buyer;
tradeFrom: the buyer obtains the amount of the false invoice/the seller issues the amount of the false invoice with 100%;
tradeTo: the buyer obtains the amount of the false invoice/the total purchase amount of the seller 100%;
control type: the personnel control the type of the enterprise, wherein 1 is a legal person, 2 is a financial person, 4 is a tax clerk, 8 is a ticket purchaser, 16 is an investor, and the combination of multiple functions is the addition of numbers.
Further, information required by algorithm construction is obtained from the initial association network, secondary integration is carried out, and a data format required by the algorithm is constructed. Specifically, in the initial association network, with the invoice virtual-open enterprise as a starting point, carrying out breadth-first traversal (BFS) on the graph to find out relationship data of an upstream layer and a downstream layer; and (4) with the personnel as a starting point, performing breadth-first traversal (BFS) of the graph to find the relationship data of a downstream layer, thereby completing the acquisition of the initial associated enterprise information and the initial associated personnel information.
(2) Infection degree calculation process:
specifically, all edges are traversed, and the infection degree of the downstream nodes is calculated for the edges meeting the conditions that the from is an invoice virtual-open enterprise and the to is a non-invoice virtual-open initial associated enterprise. Here, it should be noted that only the invoice utility and the downstream node have a utility transaction relationship, and the infection degree can be calculated. In addition, the infection degree is accumulated when a plurality of invoice virtual-open enterprises infect the same enterprise node.
Wherein, the infection degree calculation formula of the downstream non-invoice virtual-open enterprise is as follows:
sendiwight threshold + tradetto (1-threshold);
the threshold is a parameter threshold preset based on business logic, and is set to be 0.5 in the application; tradeFrom accounts for the amount of the false invoices obtained by the purchaser to the amount of the false invoices made by the seller, and tradeTo accounts for the amount of the false invoices made by the purchaser to the total amount purchased by the purchaser.
The mode of calculation of the infectivity is described by way of example below:
virtually opening an enterprise A:
{“sensi_weight”:1}；
the transaction relationship is as follows:
{“tradeFrom”:0.3,“tradeTo”:0.6}；
opening a downstream enterprise B:
{“sensiWeight”:？}
then the infectivity of B is: 0.3 × 0.5+0.6 × 0.5 ═ 0.45.
(3) And (3) calculating intimacy:
specifically, because the affinity is an attribute of the person node-associated enterprise, all the person nodes may be traversed first, and all the enterprise nodes downstream are traversed in the initial association network with the person nodes as a starting point, where a depth is set to 1, that is, the related information of the enterprise at the downstream layer is traversed. Wherein, the intimacy degree is updated for non-false-invoice enterprises in downstream enterprise nodes, and the cumulative intimacy degree associated with a plurality of invoice false-invoice enterprises for the same enterprise node is provided with personnel.
First, an affinity corresponding value is defined for the controlType attribute in the EDGE table as follows:
scores＝{‘1’:0.5,‘2’:0.2,‘4’:0.1,‘8’:0.1,‘16’:0.5,……}；
the intimacy calculation formula of the non-virtual open enterprises in the associated enterprises is as follows:
intiWeight＝scores[controlType0]+scores[controlType1]；
wherein, the controlType0 is the controlType between the personnel and the invoice virtual company, and the controlType1 is the controlType between the same personnel and the invoice virtual company.
The mode of calculation of the infectivity is described by way of example below:
a virtual enterprise A, a person M and a person downstream enterprise C; wherein,
the control relation between M and A is as follows: if the person is both a legal person and a responsible person, then { "controlType": 17 };
the control relation between M and C is as follows: the person who acts as a legal person, then { "controlType": 1 };
then, staff downstream Business C { "intiWeight":? };
then 17:0.5+0.5 ═ 1; 1: 0.5; the intimacy of C is then: 1+0.5 ═ 1.5.
Further, node filtering is performed based on a preset infection threshold and a preset intimacy threshold, and in the application, the preset infection threshold is set to be 0.35, and the preset intimacy threshold is set to be 0.6. Therefore, the nodes with the infection degrees larger than 0.35 and the nodes with the intimacy larger than 0.6 can be extracted, and the intermediate associated enterprise information and the intermediate associated personnel information are obtained.
(4) Constructing a big associated network graph:
in the step, the purpose is to reconstruct an association network big graph which clearly represents the relationship information among the invoice virtual-open enterprises, the intermediate association enterprises and the intermediate association personnel according to the filtered nodes and edges.
(5) And identifying the invoice false-open party:
calculating the association network big graph by adopting a connected community division algorithm, and filtering once to obtain a connected subgraph set with the number of enterprise nodes being more than 3 and the degree of departure of personnel nodes being more than 2; and further traversing each connected subgraph meeting the preliminary conditions in the set again based on preset service rules, judging whether the connected subgraph meets a mode that a personnel control relationship is primary and a transaction relationship is secondary, obtaining the connected subgraphs with the number of enterprise nodes more than 3 and the output degree of the personnel nodes more than 2 by secondary filtering, and marking the enterprises and the personnel involved in the connected subgraphs as puzzles, thereby obtaining the associated subgraphs and realizing the invoice false-open and puzzles identification.
(6) Verifying the invoice virtual open group and analyzing the operation mode:
specifically, the invoice virtual-open gang is subjected to service verification according to actual service rules, so that common group operation modes such as a triangular loop and a quadrilateral loop are obtained, and therefore, pattern matching can be performed in an initial association network according to the modes, and a new qualified recognition result is obtained. Aiming at the invoice false invoice issuing behavior, the corresponding business rule can be that the false invoice issuing chain is not too long, so that the false invoice issuing needs to ensure fund backflow.
(7) And (3) feedback and parameter adjustment:
specifically, according to the verification result, algorithm feedback adjustment is performed, which mainly adjusts a preset infection threshold (values such as 0.1,0.3,0.6 and 0.9), a preset intimacy threshold (values such as 0.2,0.35,0.45 and 0.7), a node number K (values such as 3,4 and 7), a mode of a connected relationship (types such as triangle, quadrangle and pentagon) and the like.
The behavior evaluation method provided by the embodiment of the application aims at the invoice virtual-open behavior, and calculates the infection degree and the affinity density of related enterprises and personnel according to the related data information of the enterprises with the invoice virtual-open behavior, thereby completing the construction of a correlation network large graph, wherein the correlation network large graph comprises the related enterprises and personnel which have a certain correlation relation with the enterprises with the invoice virtual-open behavior, further, the correlation degree screening is carried out on the correlation network large graph, and the related enterprises and personnel which are also possible to have the invoice virtual-open behavior, namely the correlation enterprises and the correlation personnel, can be identified and obtained, therefore, the related enterprises and personnel can be excavated out based on the enterprises with the invoice virtual-open behavior of one user, the identification of the invoice virtual-open group is realized, and the problem that the invoice virtual-open behavior exists in the related auditing departments is convenient for purposefully discovering the enterprises with the invoice virtual-open behavior, the system can be effectively supervised, and has higher real-time performance and pertinence; in addition, the identification of the invoice false open and group is realized based on the computer technology, and compared with the method for carrying out classification identification on individuals through manual position checking in the prior art, the method has higher identification efficiency.
To solve the above problem, please refer to fig. 4, fig. 4 is a schematic structural diagram of a behavior evaluation device provided in the present application, where the behavior evaluation device may include:
the information acquisition module 10 is used for acquiring target data information of a target enterprise; wherein the target enterprise is an enterprise with a known behavior type;
the information extraction module 20 is configured to perform information extraction on the target data information to obtain initial associated enterprise information and initial associated personnel information;
the infection degree calculation module 30 is configured to perform infection degree calculation on the initial associated enterprise according to the initial associated enterprise information, and extract intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information;
the intimacy degree calculating module 40 is used for performing intimacy degree calculation on the initial associated personnel according to the initial associated personnel information and extracting intermediate associated personnel information meeting a preset intimacy degree threshold value condition from the initial associated personnel information;
the network construction module 50 is used for constructing a correlation network large graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information;
the network screening module 60 is used for screening the association degree of the associated network large graph to obtain associated enterprises and associated personnel;
and the behavior marking module 70 is used for marking the associated enterprises and the associated personnel as the behavior types corresponding to the target enterprises.
As a preferred embodiment, the information extraction module 20 may include:
the information extraction unit is used for acquiring all related enterprise information and all related personnel information according to the target data information;
the information integration unit is used for integrating all related enterprise information and related personnel information to generate a two-dimensional wide table;
the information processing unit is used for importing the two-dimensional wide table into an ArangoDB database for processing to obtain an initial correlation network;
and the information extraction unit is used for extracting the initial associated enterprise information and the initial associated personnel information in a preset range of the initial associated network.
As a preferred embodiment, the information integration unit may be specifically configured to construct a triple transaction edge of the target enterprise-relationship-related enterprise according to all related enterprise information; constructing a first triple control edge of the related personnel-relationship-target enterprise according to all related personnel information; constructing a second triple control edge of the related personnel-relationship-related enterprise according to the triple transaction edge and the first triple control edge; and integrating the triple transaction edge, the first triple control edge and the second triple control edge to obtain the two-dimensional width table.
As a preferred embodiment, the information extracting unit may be specifically configured to extract the initial associated enterprise information and the initial associated person information within a preset range of the initial associated network by using a breadth-first search algorithm.
As a preferred embodiment, the network screening module 60 may include:
the network computing unit is used for computing the associated network big graph by utilizing a connected community division algorithm to obtain a connected subgraph set;
the network screening unit is used for screening all connected subgraphs in the connected subgraph set through a preset service rule to obtain associated subgraphs;
and the network extraction unit is used for extracting the associated enterprises and the associated persons from the associated subgraph.
As a preferred embodiment, the behavior evaluation device may further include:
the pattern analysis module is used for carrying out service verification on the associated subgraph to obtain a verification result; and determining a group operation mode according to a verification result by combining the preset connectivity mode.
As a preferred embodiment, the behavior evaluation device may further include:
the parameter optimization module is used for feeding back and adjusting parameters according to the verification result; the adjusting parameters comprise a preset infection threshold, a preset intimacy threshold, a preset range of an initial association network, a preset service rule and a preset communication relation mode.
For the introduction of the apparatus provided in the present application, please refer to the above method embodiments, which are not described herein again.
To solve the above problem, please refer to fig. 5, fig. 5 is a schematic structural diagram of a behavior evaluation device provided in the present application, where the behavior evaluation device may include:
a memory 11 for storing a computer program;
a processor 12 for implementing the following steps when executing the computer program:
acquiring target data information of a target enterprise; wherein the target enterprise is an enterprise with a known behavior type; extracting information of the target data information to obtain initial associated enterprise information and initial associated personnel information; calculating the infection degree of the initial associated enterprise according to the initial associated enterprise information, and extracting intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information; performing intimacy calculation on the initial associated personnel according to the initial associated personnel information, and extracting intermediate associated personnel information meeting a preset intimacy threshold value condition from the initial associated personnel information; constructing a correlation network big graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information; screening the association degree of the association network large graph to obtain association enterprises and association personnel; and marking the associated enterprises and the associated personnel as behavior types corresponding to the target enterprises.
For the introduction of the device provided in the present application, please refer to the above method embodiment, which is not described herein again.
To solve the above problem, the present application further provides a computer-readable storage medium having a computer program stored thereon, where the computer program when executed by a processor can implement the following steps:
acquiring target data information of a target enterprise; wherein the target enterprise is an enterprise with a known behavior type; extracting information of the target data information to obtain initial associated enterprise information and initial associated personnel information; calculating the infection degree of the initial associated enterprise according to the initial associated enterprise information, and extracting intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information; performing intimacy calculation on the initial associated personnel according to the initial associated personnel information, and extracting intermediate associated personnel information meeting a preset intimacy threshold value condition from the initial associated personnel information; constructing a correlation network big graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information; screening the association degree of the association network large graph to obtain association enterprises and association personnel; and marking the associated enterprises and the associated personnel as behavior types corresponding to the target enterprises.
The computer-readable storage medium may include: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
For the introduction of the computer-readable storage medium provided in the present application, please refer to the above method embodiments, which are not described herein again.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The behavior assessment method, apparatus, device, and computer-readable storage medium provided by the present application are described in detail above. The principles and embodiments of the present application are explained herein using specific examples, which are provided only to help understand the method and the core idea of the present application. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and these improvements and modifications also fall into the elements of the protection scope of the claims of the present application.
Claims (10)
1. A method of behavioral assessment, comprising:
acquiring target data information of a target enterprise; wherein the target enterprise is an enterprise of a known behavior type;
extracting information from the target data information to obtain initial associated enterprise information and initial associated personnel information;
calculating the infection degree of the initial associated enterprise according to the initial associated enterprise information, and extracting intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information;
performing intimacy calculation on initial associated personnel according to the initial associated personnel information, and extracting intermediate associated personnel information meeting a preset intimacy threshold value condition from the initial associated personnel information;
constructing a correlation network big graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information;
screening the association degree of the association network large graph to obtain an association enterprise and an association person;
and marking the associated enterprises and the associated personnel as behavior types corresponding to the target enterprises.
2. The behavior assessment method according to claim 1, wherein the extracting information from the target data information to obtain initial associated business information and initial associated personnel information comprises:
acquiring all related enterprise information and all related personnel information according to the target data information;
integrating all the related enterprise information and the related personnel information to generate a two-dimensional broad table;
importing the two-dimensional wide table into an ArangoDB graph database for processing to obtain an initial association network;
and extracting the initial associated enterprise information and the initial associated personnel information in a preset range of the initial associated network.
3. The method for behavioral assessment according to claim 2, wherein said integrating all of said related business information and said related personnel information to generate a two-dimensional broad table comprises:
constructing a triple transaction edge of a target enterprise-relation-related enterprise according to all the related enterprise information;
constructing a first triple control edge of the related personnel-relationship-target enterprise according to all the related personnel information;
constructing a second triple control edge of the related personnel-relationship-related enterprise according to the triple transaction edge and the first triple control edge;
and integrating the triple transaction edge, the first triple control edge and the second triple control edge to obtain the two-dimensional wide table.
4. The behavior assessment method according to claim 2, wherein the extracting the initial associated business information and the initial associated personnel information within a preset range of the initial associated network comprises:
and extracting the initial associated enterprise information and the initial associated personnel information in a preset range of the initial associated network by using a breadth-first search algorithm.
5. The behavior assessment method according to claim 4, wherein the screening of the association degree of the association network large graph to obtain the association enterprises and the association personnel comprises:
calculating the associated network big graph by using a connected community division algorithm to obtain a connected subgraph set;
screening all connected subgraphs in the connected subgraph set through a preset service rule to obtain associated subgraphs;
and extracting the associated enterprises and the associated persons from the associated subgraph.
6. The behavior assessment method according to any one of claims 1 to 5, further comprising:
performing service verification on the associated subgraph to obtain a verification result;
and determining a group operation mode according to the verification result by combining a preset connectivity mode.
7. The behavior assessment method of claim 6, further comprising:
feeding back and adjusting parameters according to the verification result; the adjustment parameters comprise the preset infection threshold, the preset intimacy threshold, the preset range of the initial association network, the preset business rule and the preset communication relation mode.
8. A behavior evaluation device, comprising:
the information acquisition module is used for acquiring target data information of a target enterprise; wherein the target enterprise is an enterprise of a known behavior type;
the information extraction module is used for extracting information from the target data information to obtain initial associated enterprise information and initial associated personnel information;
the infection degree calculation module is used for calculating the infection degree of the initial associated enterprise according to the initial associated enterprise information and extracting intermediate associated enterprise information meeting a preset infection degree threshold condition from the initial associated enterprise information;
the intimacy degree calculation module is used for carrying out intimacy degree calculation on initial associated personnel according to the initial associated personnel information and extracting intermediate associated personnel information meeting a preset intimacy degree threshold value condition from the initial associated personnel information;
the network construction module is used for constructing a correlation network big graph according to the intermediate correlation enterprise information and the intermediate correlation personnel information;
the network screening module is used for screening the association degree of the associated network large graph to obtain associated enterprises and associated personnel;
and the behavior marking module is used for marking the associated enterprises and the associated personnel as behavior types corresponding to the target enterprises.
9. A behavior evaluation device characterized by comprising:
a memory for storing a computer program;
a processor for implementing the steps of the behavior assessment method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by a processor, carries out the steps of the behavior assessment method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811550706.XA CN109635007B (en) | 2018-12-18 | 2018-12-18 | Behavior evaluation method and device and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811550706.XA CN109635007B (en) | 2018-12-18 | 2018-12-18 | Behavior evaluation method and device and related equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109635007A true CN109635007A (en) | 2019-04-16 |
| CN109635007B CN109635007B (en) | 2020-10-27 |
Family
ID=66075146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811550706.XA Active CN109635007B (en) | 2018-12-18 | 2018-12-18 | Behavior evaluation method and device and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109635007B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110399431A (en) * | 2019-07-16 | 2019-11-01 | 阿里巴巴集团控股有限公司 | A kind of incidence relation construction method, device and equipment |
| CN111241347A (en) * | 2019-12-30 | 2020-06-05 | 北京邮电大学 | Graph database creating method, enterprise data query method and device |
| CN112241914A (en) * | 2020-09-30 | 2021-01-19 | 航天信息股份有限公司 | Enterprise evaluation method and device, storage medium and electronic equipment |
| CN112287039A (en) * | 2020-10-30 | 2021-01-29 | 税友软件集团股份有限公司 | Group partner identification method and related device |
| CN112418652A (en) * | 2020-11-19 | 2021-02-26 | 税友软件集团股份有限公司 | Risk identification method and related device |
| CN112905649A (en) * | 2021-02-08 | 2021-06-04 | 珠海金山网络游戏科技有限公司 | Query method and device |
| CN113313433A (en) * | 2021-07-13 | 2021-08-27 | 平安科技（深圳）有限公司 | Conference resource allocation method based on knowledge graph and related equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020128910A1 (en) * | 2001-01-10 | 2002-09-12 | Takuya Sakuma | Business supporting system and business supporting method |
| US20170270426A1 (en) * | 2016-03-17 | 2017-09-21 | VEDA Data Solutions LLC | Performing regression analysis on personal data records |
| CN107918905A (en) * | 2017-11-22 | 2018-04-17 | 阿里巴巴集团控股有限公司 | Abnormal transaction identification method, apparatus and server |
| CN108197903A (en) * | 2018-02-02 | 2018-06-22 | 金蝶软件（中国）有限公司 | A kind of relation information processing method and processing device in enterprise |
-
2018
- 2018-12-18 CN CN201811550706.XA patent/CN109635007B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020128910A1 (en) * | 2001-01-10 | 2002-09-12 | Takuya Sakuma | Business supporting system and business supporting method |
| US20170270426A1 (en) * | 2016-03-17 | 2017-09-21 | VEDA Data Solutions LLC | Performing regression analysis on personal data records |
| CN107918905A (en) * | 2017-11-22 | 2018-04-17 | 阿里巴巴集团控股有限公司 | Abnormal transaction identification method, apparatus and server |
| CN108197903A (en) * | 2018-02-02 | 2018-06-22 | 金蝶软件（中国）有限公司 | A kind of relation information processing method and processing device in enterprise |
Non-Patent Citations (1)
| Title |
|---|
| 王久铱: "《基于聚类的税务稽查选案方法及其系统的研究》", 《中国优秀硕士学位论文全文数据库（电子期刊）》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110399431A (en) * | 2019-07-16 | 2019-11-01 | 阿里巴巴集团控股有限公司 | A kind of incidence relation construction method, device and equipment |
| CN111241347A (en) * | 2019-12-30 | 2020-06-05 | 北京邮电大学 | Graph database creating method, enterprise data query method and device |
| CN112241914A (en) * | 2020-09-30 | 2021-01-19 | 航天信息股份有限公司 | Enterprise evaluation method and device, storage medium and electronic equipment |
| CN112287039A (en) * | 2020-10-30 | 2021-01-29 | 税友软件集团股份有限公司 | Group partner identification method and related device |
| CN112418652A (en) * | 2020-11-19 | 2021-02-26 | 税友软件集团股份有限公司 | Risk identification method and related device |
| CN112418652B (en) * | 2020-11-19 | 2024-01-30 | 税友软件集团股份有限公司 | Risk identification method and related device |
| CN112905649A (en) * | 2021-02-08 | 2021-06-04 | 珠海金山网络游戏科技有限公司 | Query method and device |
| CN113313433A (en) * | 2021-07-13 | 2021-08-27 | 平安科技（深圳）有限公司 | Conference resource allocation method based on knowledge graph and related equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109635007B (en) | 2020-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109635007B (en) | Behavior evaluation method and device and related equipment | |
| CN108665159A (en) | A kind of methods of risk assessment, device, terminal device and storage medium | |
| CN110647590A (en) | Target community data identification method and related device | |
| WO2021254027A1 (en) | Method and apparatus for identifying suspicious community, and storage medium and computer device | |
| Franceschetti et al. | Do bankrupt companies manipulate earnings more than the non-bankrupt ones? | |
| WO2017133456A1 (en) | Method and device for determining risk evaluation parameter | |
| Papik et al. | Detection models for unintentional financial restatements | |
| CN104915879A (en) | Social relationship mining method and device based on financial data | |
| WO2020134213A1 (en) | Method and system for querying abnormal financial data on basis of knowledge map | |
| CN111861595A (en) | Cyclic invoicing risk identification method based on knowledge graph | |
| CN104424613A (en) | Value added tax invoice monitoring method and system thereof | |
| CN111833182B (en) | Method and device for identifying risk object | |
| CN111090780A (en) | Method and device for determining suspicious transaction information, storage medium and electronic equipment | |
| Da Silva et al. | Selecting audit samples using Benford's Law | |
| CN108572988A (en) | A kind of house property assessment data creation method and device | |
| CN112612813A (en) | Test data generation method and device | |
| CN112287039A (en) | Group partner identification method and related device | |
| CN113888278A (en) | Data analysis method and device based on enterprise credit line analysis model | |
| Garin et al. | Machine learning in classifying bitcoin addresses | |
| CN113240259A (en) | Method and system for generating rule policy group and electronic equipment | |
| CN112950290A (en) | Mining method and device for economic dependence clients, storage medium and electronic equipment | |
| CN109636627B (en) | Insurance product management method, device, medium and electronic equipment based on block chain | |
| CN116051272A (en) | Enterprise risk analysis method and related equipment | |
| Dhurandhar et al. | Robust system for identifying procurement fraud | |
| CN114626940A (en) | Data analysis method and device and electronic equipment |
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 |