A method and device for mining tax risk enterprises

Abstract

The present application relates to the field of big data mining technology, and in particular to a method and device for mining tax risk enterprises. The method includes: obtaining the enterprise information of blacklisted enterprises, and obtaining the ticket flow relationship between the blacklisted enterprises and other enterprises; based on the enterprise information of the blacklisted enterprises and the ticket flow relationship between the blacklisted enterprises and other enterprises, constructing a directed graph structure with the blacklisted enterprises and the other enterprises as nodes, and the ticket flow relationship between the blacklisted enterprises and the other enterprises as edges; based on the directed graph structure, calculating the correlation between each node, and according to the correlation between each node, performing clustering processing on the associated enterprises for each node to obtain at least one enterprise set; based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, respectively calculating the tax risk value of each enterprise set.

Description

Technical Field

[0001] The present application relates to the field of big data mining technology, and in particular to a method and device for mining tax risk enterprises. Background Art

[0002] With the rapid development of my country's economy, tax supervision has become more and more important. Tax supervision departments need to accurately crack down on companies that evade taxes, comprehensively investigate tax risk companies, and uncover risky trading gangs. Therefore, how to discover corporate trading gangs has become a very valuable issue.

[0003] The current method of finding risky corporate trading groups is still limited to identifying all upstream and downstream companies involved in blacklisted companies based on the subjective experience of tax officials and the objective rules of corporate transactions. This traditional method is not only inefficient and difficult to guarantee accuracy, but also has a large amount of information on the face value data, and manual review consumes a lot of manpower costs.

[0004] Therefore, there is an urgent need for a fast, intelligent, accurate and comprehensive method for mining risk groups to assist tax officials in screening and assessing risky enterprises and combating corporate tax evasion. Summary of the invention

[0005] The present application provides a tax risk enterprise mining method and device to solve the problems of large workload and low accuracy in the prior art of tax risk enterprise mining.

[0006] In a first aspect, the present application provides a method for mining tax risk enterprises, the method comprising:

[0007] Obtaining the enterprise information of the blacklisted enterprise, and obtaining the ticket flow relationship between the blacklisted enterprise and other enterprises;

[0008] Based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, a directed graph structure is constructed with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges;

[0009] Based on the directed graph structure, calculating the association degree between each node, and performing clustering processing on the associated enterprises of each node according to the association degree between each node, so as to obtain at least one enterprise set;

[0010] Based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, the tax risk value of each enterprise set is calculated separately.

[0011] Optionally, the corporate information of an enterprise includes the taxpayer identification number and the taxpayer name; the ticket flow relationship includes at least the supplier's corporate information, the seller's corporate information, the transaction amount, and the transaction time.

[0012] Optionally, based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, the step of constructing a directed graph structure with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges includes:

[0013] Pre-process the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises to obtain an enterprise list with one-to-one correspondence between the taxpayer identification number and the taxpayer name;

[0014] Adding the enterprises in the enterprise list as nodes to the graph database;

[0015] The ticket flow relationships involved in all enterprises in the enterprise list are added as edges to the graph database, wherein the edge pointing to a node is the input edge of the node, and the edge pointed by a node is the output edge of the node. The attribute information of the edge includes: transaction amount and transaction time;

[0016] A directed graph structure is constructed by each node and the input edge and / or output edge corresponding to each node.

[0017] Optionally, based on the directed graph structure, calculating the association degree between each node, and performing clustering processing on the nodes according to the association degree between each node, to obtain at least one enterprise set comprises:

[0018] Based on the directed graph structure, the out-degree and / or in-degree contribution between each node is calculated, and the associated enterprises of each node are clustered according to the out-degree and / or in-degree contribution to obtain at least one enterprise set.

[0019] Optionally, based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, the step of calculating the tax risk value of each enterprise set includes:

[0020] The tax risk value of the target enterprise set is calculated using the following formula:

[0021] RiskValue=C+100*P

[0022] Among them, RiskValue is the tax risk value of the target enterprise set, C is the number of blacklisted enterprises included in the target enterprise set, and P is the ratio of the number of blacklisted enterprises included in the target enterprise set to the total number of enterprises.

[0023] In a second aspect, the present application provides a tax risk enterprise mining device, the device comprising:

[0024] An acquisition unit, used to acquire enterprise information of a blacklist enterprise and acquire a ticket flow relationship between the blacklist enterprise and other enterprises;

[0025] A construction unit, configured to construct, based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and the other enterprises, a directed graph structure with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges;

[0026] A clustering unit, configured to calculate the association degree between each node based on the directed graph structure, and perform clustering processing of associated enterprises on each node according to the association degree between each node to obtain at least one enterprise set;

[0027] The calculation unit is used to calculate the tax risk value of each enterprise set based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set.

[0028] Optionally, the corporate information of an enterprise includes the taxpayer identification number and the taxpayer name; the ticket flow relationship includes at least the supplier's corporate information, the seller's corporate information, the transaction amount, and the transaction time.

[0029] Optionally, based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, when a directed graph structure is constructed with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges, the construction unit is specifically used to:

[0030] Pre-process the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises to obtain an enterprise list with one-to-one correspondence between the taxpayer identification number and the taxpayer name;

[0031] Adding the enterprises in the enterprise list as nodes to the graph database;

[0032] The ticket flow relationships involved in all enterprises in the enterprise list are added as edges to the graph database, wherein the edge pointing to a node is the input edge of the node, and the edge pointed by a node is the output edge of the node. The attribute information of the edge includes: transaction amount and transaction time;

[0033] A directed graph structure is constructed by each node and the input edge and / or output edge corresponding to each node.

[0034] Optionally, based on the directed graph structure, the association degree between each node is calculated, and when the nodes are clustered into associated enterprises according to the association degree between each node to obtain at least one enterprise set, the clustering unit is specifically used to:

[0035] Based on the directed graph structure, calculating the out-degree and / or in-degree contribution between each node;

[0036] According to the out-degree and / or in-degree contribution, the associated enterprises of each node are clustered to obtain at least one enterprise set.

[0037] Optionally, when calculating the tax risk value of each enterprise set based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, the calculation unit is specifically used to:

[0038] The tax risk value of the target enterprise set is calculated using the following formula:

[0039] RiskValue=C+100*P

[0040] Among them, RiskValue is the tax risk value of the target enterprise set, C is the number of blacklisted enterprises included in the target enterprise set, and P is the ratio of the number of blacklisted enterprises included in the target enterprise set to the total number of enterprises.

[0041] In a third aspect, the present application provides a tax risk enterprise mining device, which includes:

[0042] A memory for storing program instructions;

[0043] The processor is used to call the program instructions stored in the memory, and execute the steps of the method as described in any one of the first aspects above according to the obtained program instructions.

[0044] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute the steps of the method described in any one of the first aspects above.

[0045] In summary, the tax risk enterprise mining method provided in the embodiment of the present application obtains the enterprise information of the blacklisted enterprise, and obtains the ticket flow relationship between the blacklisted enterprise and other enterprises; based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, a directed graph structure is constructed with the blacklisted enterprise and the other enterprises as nodes, and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges; based on the directed graph structure, the correlation between each node is calculated, and according to the correlation between each node, the nodes are clustered into associated enterprises to obtain at least one enterprise set; based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, the tax risk value of each enterprise set is calculated respectively.

[0046] The tax risk enterprise mining method provided in the embodiment of the present application is adopted, based on the known blacklist enterprises and the ticket flow relationship between the blacklist enterprises and other enterprises, each enterprise is clustered, and each enterprise is divided into at least one enterprise cluster according to the correlation between each enterprise, and the tax risk value of each enterprise cluster is counted according to the total number of enterprises included in each enterprise cluster and the number of known blacklist enterprises, and then according to the tax risk value of each enterprise cluster, the tax staff is reminded to conduct specific verification of the enterprise cluster with higher tax risk value. This enables the tax staff to conduct targeted verification of the enterprise cluster with high tax risk value, improves the efficiency of tax risk enterprise mining, and improves the accuracy of tax risk enterprise mining. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments of the present application or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings of the embodiments of the present application.

[0048] Figure 1 A detailed flow chart of a tax risk enterprise mining method provided in an embodiment of the present application;

[0049] Figure 2 A schematic diagram of the proportion of out-degree contribution provided in an embodiment of the present application;

[0050] Figure 3 A schematic diagram of the proportion of in-degree contribution provided in an embodiment of the present application;

[0051] Figure 4 A schematic diagram of the structure of a tax risk enterprise mining device provided in an embodiment of the present application;

[0052] Figure 5 A schematic diagram of the structure of another tax risk enterprise mining device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, rather than limiting the present application. The singular forms of "a", "said" and "the" used in the present application and claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to any or all possible combinations of one or more associated listed items.

[0054] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in the embodiments of the present application, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, in addition, the word "if" used may be interpreted as "at..." or "when..." or "in response to determination".

[0055] For example, see Figure 1 FIG. 1 is a detailed flow chart of a tax risk enterprise mining method provided in an embodiment of the present application, and the method includes the following steps:

[0056] Step 100: Obtain the enterprise information of the blacklist enterprise, and obtain the ticket flow relationship between the blacklist enterprise and other enterprises.

[0057] In an embodiment of the present application, the corporate information of an enterprise includes a taxpayer identification number and a taxpayer name; the ticket flow relationship includes at least the supplier's corporate information, the seller's corporate information, the transaction amount, and the transaction time.

[0058] In actual applications, the information of known blacklisted enterprises is collected in advance, including at least the taxpayer identification number and the taxpayer name. Transactions involving blacklisted enterprises are collected and recorded, including the seller's taxpayer identification number, seller's taxpayer name, supplier's taxpayer identification number, supplier's taxpayer name, transaction amount, transaction time, etc. in the invoice information. For example, the specific contents of the invoice information are shown in Table 1.

[0059] Table 1

[0060]

[0061] Step 110: Based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, a directed graph structure is constructed with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges.

[0062] In the embodiment of the present application, based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, a directed graph structure is constructed with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges. A preferred implementation method is:

[0063] Preprocess the enterprise information of the blacklisted enterprises and the ticket flow relationship between the blacklisted enterprises and other enterprises to obtain an enterprise list with one-to-one correspondence between taxpayer identification numbers and taxpayer names; add the enterprises in the enterprise list as nodes to the graph database; add the ticket flow relationships involving all enterprises in the enterprise list as edges to the graph database, wherein an edge pointing to a node is regarded as an input edge of the node, and an edge pointed out by a node is regarded as an output edge of the node, and the attribute information of the edge includes: transaction amount, transaction time; a directed graph structure is constructed by each node and the input edge and / or output edge corresponding to each node.

[0064] Specifically, based on the collected data set and the ticket flow relationship between enterprises, a directed graph structure is constructed and imported into the JanusGraph database. The main steps of the directed graph structure construction process include node construction and edge construction.

[0065] The node construction process is as follows: pre-process the collected data (corporate information of blacklisted companies and the ticket flow relationship between blacklisted companies and other companies), screen all taxpayer names and taxpayer identification numbers, handle data errors where one name corresponds to multiple tax numbers or one tax number corresponds to multiple names, output a one-to-one list of companies with taxpayer identification numbers and taxpayer names, add the companies in the company list as nodes to the graph database, and mark the known blacklisted companies in the list with the label black_nsr, and the rest of the companies with nsr. The attribute information of the node includes: taxpayer name, taxpayer identification number, and other relevant business information.

[0066] The edge construction process is as follows: add the ticket flow relationships of all enterprises in the enterprise list as edges to the graph database, the edge pointing to the enterprise (node) is the input edge, the label is input, and the edge pointed by the enterprise (node) is the output edge, the label is output. The attribute information of the edge includes: transaction amount, transaction time, etc.

[0067] Step 120: Based on the directed graph structure, the association degree between each node is calculated, and according to the association degree between each node, the nodes are clustered into related enterprises to obtain at least one enterprise set.

[0068] In the embodiment of the present application, based on the directed graph structure, the association degree between each node is calculated, and according to the association degree between each node, the nodes are clustered into associated enterprises to obtain at least one enterprise set. A preferred implementation method is:

[0069] Based on the directed graph structure, the out-degree and / or in-degree contribution between each node is calculated, and the associated enterprises of each node are clustered according to the out-degree and / or in-degree contribution to obtain at least one enterprise set.

[0070] In practical applications, based on the constructed (directed graph structure) purchase and sales relationship graph network, the PeerPressure community discovery algorithm is used to divide the enterprise's transaction community (enterprise set) and output the list of enterprises in each community.

[0071] The PeerPressure community (in the embodiment of the present application, the community is taken as an example of a collection of enterprises) division algorithm specifically includes:

[0072] Step 1: Preprocess the data to obtain the taxpayer entity and the ticket flow relationship between entities, and construct a graph network;

[0073] Step 2: Process the eigenvalues ​​and calculate the proportion of each vertex (node) to its neighboring vertices' out-degree (output) and the vertex that contributes the most to its in-degree (input).

[0074] For example, Figure 2 As shown, it is a schematic diagram of the proportion of out-degree contribution provided in an embodiment of the present application. For example, enterprise A has 3 out-edges connecting enterprises B, C and D respectively, then the proportion of the out-degree of vertex A to its adjacent vertices is 1 / (1+3)=0.25.

[0075] For example, see Figure 3 As shown, it is a schematic diagram of in-degree contribution provided in an embodiment of the present application. For example, enterprise A has two incoming edges connecting enterprises B and C respectively, where the out-degree ratio of vertex B to its adjacent vertices is 0.5, and that of vertex C is 0.33. Then vertex B is the vertex with the largest in-degree contribution to vertex A.

[0076] Step 3: Approximately cluster the nodes. If a node A has only one vertex B that contributes the most to its in-degree, then A and B are grouped into one category.

[0077] Step 4: Repeat steps 2 and 3, recalculate the eigenvalues ​​of each vertex, and perform approximate clustering until the clustering result no longer changes.

[0078] Step 130: Based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, the tax risk value of each enterprise set is calculated respectively.

[0079] In the embodiment of the present application, when the tax risk value of each enterprise set is calculated based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, a preferred implementation method is:

[0080] The tax risk value of the target enterprise set is calculated using the following formula. The target enterprise set is any set of clustered enterprises:

[0081] RiskValue=C+100*P

[0082] Among them, RiskValue is the tax risk value of the target enterprise set, C is the number of blacklisted enterprises included in the target enterprise set, and P is the ratio of the number of blacklisted enterprises included in the target enterprise set to the total number of enterprises.

[0083] Specifically, according to the results of community division, the list of enterprises in each community is counted, the number C and proportion P of known blacklisted enterprises in each community are output, and the risk value of each community is obtained through the community risk value calculation formula. The calculation formula is as follows:

[0084] RiskValue=C+100*P

[0085] For example, there are 100 enterprises in community 1, of which 20 are known blacklisted enterprises. Then the risk value of the community is 20+100*0.2=40.

[0086] Furthermore, all community risk values ​​obtained are normalized and risk thresholds are set. Communities exceeding the thresholds are defined as risky trading communities, and enterprises within the communities are specifically verified by tax officials.

[0087] Based on the same inventive concept as the above method embodiment applied to the first service host, for example, refer to Figure 4 FIG. 1 is a schematic diagram of a memory over-allocation management device provided in an embodiment of the present application. The device is applied to a first service host in a cluster. The cluster also includes several other service hosts and a management host. The device includes:

[0088] The sending unit 40 is used to send a notification to the management host when it is determined that the local memory usage is greater than or equal to a set threshold, so that the management host queries the memory usage of the other business hosts in the cluster, and determines a second business host whose memory usage meets the preset requirement based on the memory usage of the other business hosts;

[0089] The receiving unit 41 is used to receive the second service host information sent by the management host;

[0090] The application unit 42 is used to establish a remote direct memory access connection with the second service host based on the second service host information, and apply for a memory pool from the second service host.

[0091] Optionally, when applying for a memory pool from the second service host, the applying unit 42 is specifically configured to:

[0092] Sending a memory application request message to the second service host, wherein the memory application request message includes the size of the memory applied for;

[0093] A response message including physical address information of an allocated memory pool of a specified size and a credential for using the memory pool of the specified size is received from the second service host.

[0094] Optionally, the device further comprises:

[0095] The detection unit allocates the target memory of the memory pool of the specified size to any virtual machine for use when it is detected that any locally running virtual machine applies for memory, and marks the physical address corresponding to the target memory as being in use; marks the physical address corresponding to the target memory as being in unused state when it is detected that any virtual machine releases the target memory; and disconnects the remote direct memory access connection with the second business host when it is detected that the idle time of the memory pool of the specified size is greater than or equal to a preset time, so that the second business host reclaims the memory pool of the specified size.

[0096] Optionally,

[0097] When it is determined that the local memory usage is greater than or equal to the set threshold, the service host is marked as not sharing memory externally and notifies the management host, wherein a service host does not provide memory usage services externally under the condition of not sharing memory externally.

[0098] Based on the same inventive concept as the above-mentioned embodiment of the invention, for example, refer to Figure 4 FIG. 1 is a schematic diagram of a tax risk enterprise mining device provided by the present application, wherein the device comprises:

[0099] The acquisition unit 40 is used to acquire the enterprise information of the blacklist enterprise and acquire the ticket flow relationship between the blacklist enterprise and other enterprises;

[0100] A construction unit 41 is used to construct a directed graph structure with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and the other enterprises;

[0101] A clustering unit 42 is used to calculate the association degree between each node based on the directed graph structure, and perform clustering processing of the associated enterprises on each node according to the association degree between each node to obtain at least one enterprise set;

[0102] The calculation unit 43 is used to calculate the tax risk value of each enterprise set based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set.

[0103] Optionally, the corporate information of an enterprise includes the taxpayer identification number and the taxpayer name; the ticket flow relationship includes at least the supplier's corporate information, the seller's corporate information, the transaction amount, and the transaction time.

[0104] Optionally, based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, when a directed graph structure is constructed with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges, the construction unit 41 is specifically used to:

[0105] Pre-process the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises to obtain an enterprise list with one-to-one correspondence between the taxpayer identification number and the taxpayer name;

[0106] Adding the enterprises in the enterprise list as nodes to the graph database;

[0107] The ticket flow relationships involved in all enterprises in the enterprise list are added as edges to the graph database, wherein the edge pointing to a node is the input edge of the node, and the edge pointed by a node is the output edge of the node. The attribute information of the edge includes: transaction amount and transaction time;

[0108] A directed graph structure is constructed by each node and the input edge and / or output edge corresponding to each node.

[0109] Optionally, based on the directed graph structure, the association degree between each node is calculated, and when the nodes are clustered into associated enterprises according to the association degree between each node to obtain at least one enterprise set, the clustering unit 42 is specifically used to:

[0110] Based on the directed graph structure, calculating the out-degree and / or in-degree contribution between each node;

[0111] According to the out-degree and / or in-degree contribution, the associated enterprises of each node are clustered to obtain at least one enterprise set.

[0112] Optionally, when calculating the tax risk value of each enterprise set based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, the calculation unit 43 is specifically used to:

[0113] The tax risk value of the target enterprise set is calculated using the following formula:

[0114] RiskValue=C+100*P

[0115] Among them, RiskValue is the tax risk value of the target enterprise set, C is the number of blacklisted enterprises included in the target enterprise set, and P is the ratio of the number of blacklisted enterprises included in the target enterprise set to the total number of enterprises.

[0116] The above units may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital singnal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a certain unit above is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call program codes. For another example, these units may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0117] Furthermore, the tax risk enterprise mining device provided in the embodiment of the present application, from the hardware level, the hardware architecture diagram of the tax risk enterprise mining device can be seen in Figure 5 As shown, the tax risk enterprise mining device may include: a memory 50 and a processor 51,

[0118] The memory 50 is used to store program instructions; the processor 51 calls the program instructions stored in the memory 50 and executes the above method embodiment according to the obtained program instructions. The specific implementation method and technical effect are similar and will not be repeated here.

[0119] Optionally, the present application also provides a tax risk enterprise mining device, comprising at least one processing element (or chip) for executing the above method embodiment.

[0120] Optionally, the present application also provides a program product, such as a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable the computer to execute the above method embodiments.

[0121] Here, the machine-readable storage medium may be any electronic, magnetic, optical or other physical storage device that may contain or store information, such as executable instructions, data, etc. For example, the machine-readable storage medium may be: RAM (RadomAccess Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard disk drive), solid state drive, any type of storage disk (such as CD, DVD, etc.), or similar storage medium, or a combination thereof.

[0122] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which may be in the form of a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email transceiver, a game console, a tablet computer, a wearable device or a combination of any of these devices.

[0123] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0124] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the embodiments of the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0125] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0126] Moreover, these computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0127] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0128] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for mining tax risk enterprises, characterized in that: The method comprises: Obtaining the enterprise information of the blacklisted enterprise, and obtaining the ticket flow relationship between the blacklisted enterprise and other enterprises; Based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, a directed graph structure is constructed with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges; Based on the directed graph structure, calculating the association degree between each node, and performing clustering processing on the associated enterprises of each node according to the association degree between each node, so as to obtain at least one enterprise set; Based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, calculate the tax risk value of each enterprise set respectively; Based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises, the steps of constructing a directed graph structure with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges include: pre-processing the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and other enterprises to obtain an enterprise list with one-to-one correspondence between taxpayer identification numbers and taxpayer names; adding the enterprises in the enterprise list as nodes to the graph database; adding the ticket flow relationships involved in all enterprises in the enterprise list as edges to the graph database, wherein an edge pointing to a node is an input edge of the node, and an edge pointed out by a node is an output edge of the node, and the attribute information of the edge includes: transaction amount, transaction time; constructing a directed graph structure with each node and the input edge and / or output edge corresponding to each node; wherein the enterprise information of the blacklisted enterprise includes taxpayer identification number and taxpayer name; the ticket flow relationship includes at least supplier enterprise information, seller enterprise information, transaction amount, and transaction time; Among them, the enterprise information of the blacklisted enterprises and the ticket flow relationship between the blacklisted enterprises and other enterprises are pre-processed to obtain a one-to-one enterprise list of taxpayer identification numbers and taxpayer names, including: screening out all taxpayer names and taxpayer identification numbers from the collected data, processing data errors where one name corresponds to multiple tax numbers or one tax number corresponds to multiple names, and outputting a one-to-one enterprise list of taxpayer identification number-taxpayer name.

2. The method according to claim 1, characterized in that The steps of calculating the association degree between each node based on the directed graph structure, and clustering the associated enterprises of each node according to the association degree between each node to obtain at least one enterprise set include: Based on the directed graph structure, the out-degree and / or in-degree contribution between each node is calculated, and the associated enterprises of each node are clustered according to the out-degree and / or in-degree contribution to obtain at least one enterprise set.

3. The method according to any one of claims 1 to 2, characterized in that: Based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, the steps of calculating the tax risk value of each enterprise set include: The tax risk value of the target enterprise set is calculated using the following formula: RiskValue=C+100*P Among them, RiskValue is the tax risk value of the target enterprise set, C is the number of blacklisted enterprises included in the target enterprise set, and P is the ratio of the number of blacklisted enterprises included in the target enterprise set to the total number of enterprises.

4. A tax risk enterprise mining device, characterized in that: The device comprises: An acquisition unit, used to acquire enterprise information of a blacklist enterprise and acquire a ticket flow relationship between the blacklist enterprise and other enterprises; A construction unit, configured to construct, based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and the other enterprises, a directed graph structure with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges; A clustering unit, configured to calculate the association degree between each node based on the directed graph structure, and perform clustering processing of associated enterprises on each node according to the association degree between each node to obtain at least one enterprise set; A calculation unit, used to calculate the tax risk value of each enterprise set based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set; When constructing a directed graph structure with the blacklisted enterprise and the other enterprises as nodes and the ticket flow relationship between the blacklisted enterprise and the other enterprises as edges based on the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and the other enterprises, the construction unit is specifically used to: pre-process the enterprise information of the blacklisted enterprise and the ticket flow relationship between the blacklisted enterprise and the other enterprises to obtain an enterprise list with one-to-one correspondence between taxpayer identification numbers and taxpayer names; add the enterprises in the enterprise list as nodes to the graph database; add the ticket flow relationships involved in all enterprises in the enterprise list as edges to the graph database, wherein an edge pointing to a node is an input edge of the node, and an edge pointed out by a node is an output edge of the node, and the attribute information of the edge includes: transaction amount, transaction time; construct a directed graph structure with each node and the input edge and / or output edge corresponding to each node; wherein the enterprise information of the blacklisted enterprise includes the taxpayer identification number and the taxpayer name; the ticket flow relationship includes at least the supplier enterprise information, the seller enterprise information, the transaction amount, and the transaction time; Pre-process the corporate information of the blacklisted enterprises and the ticket flow relationship between the blacklisted enterprises and other enterprises to obtain a corporate list with one-to-one correspondence between taxpayer identification numbers and taxpayer names, including: screening out all taxpayer names and taxpayer identification numbers from the collected data, processing data errors where one name corresponds to multiple tax numbers or one tax number corresponds to multiple names, and outputting a one-to-one corporate list of taxpayer identification number-taxpayer name.

5. The device according to claim 4, characterized in that Based on the directed graph structure, the association degree between each node is calculated, and when the nodes are clustered into associated enterprises according to the association degree between each node to obtain at least one enterprise set, the clustering unit is specifically used for: Based on the directed graph structure, calculating the out-degree and / or in-degree contribution between each node; According to the out-degree and / or in-degree contribution, the associated enterprises of each node are clustered to obtain at least one enterprise set.

6. The device according to any one of claims 4 to 5, characterized in that: When calculating the tax risk value of each enterprise set based on the number of blacklisted enterprises and the total number of enterprises included in each enterprise set, the calculation unit is specifically used to: The tax risk value of the target enterprise set is calculated using the following formula: RiskValue=C+100*P Among them, RiskValue is the tax risk value of the target enterprise set, C is the number of blacklisted enterprises included in the target enterprise set, and P is the ratio of the number of blacklisted enterprises included in the target enterprise set to the total number of enterprises.

Images