Enterprise evaluation method, device, storage medium and electronic device
By constructing a network of multiple related enterprises and using a random walk model to assess enterprise importance, the problem of low accuracy in ranking enterprise importance in existing technologies is solved, and a more accurate assessment of enterprise importance is achieved.
Patent Information
- Application Number
- CN202011063041.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-09-30
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2040-09-30
AI Technical Summary
In existing technologies, most enterprise importance ranking algorithms in the tax industry evaluate single network data, resulting in low evaluation accuracy.
By acquiring different business data information from the tax data of the companies to be evaluated, multiple networks of related companies are established, and a random walk model is used to determine the importance weight of the companies, including the construction and network fusion processing of the enterprise-natural person, enterprise-enterprise invoice flow and enterprise-purchase and sales goods meta-network.
It improves the accuracy of enterprise importance assessment, enables the establishment of a network of related enterprises from multiple perspectives, and provides a more comprehensive ranking of enterprise importance.
Smart Images

Figure CN112241914B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of tax big data, and in particular, to an enterprise evaluation method, device, storage medium and electronic device. Background Art
[0002] With the development of tax big data, the industry has a large amount of cash settlement, and the tax bureau's supervision and management is difficult. It is very necessary to mine important enterprises from a large amount of invoice data so that the tax department can perform supervision efficiently and quickly. By mining important enterprises in the tax field, key indicators such as industry leading enterprises can be provided to support tax data analysis and decision-making. Therefore, the enterprise importance ranking method plays an important guiding role in the analysis and mining of big data in the tax industry. However, there are currently few studies on the enterprise importance ranking algorithm in the tax industry, and most of them are calculated and evaluated for single network data, which has strong limitations, making the accuracy of the assessed enterprise importance low. Summary of the invention
[0003] The purpose of the present disclosure is to provide an enterprise evaluation method, device, storage medium and electronic device to improve the accuracy of enterprise importance evaluation.
[0004] In order to achieve the above-mentioned purpose, the present disclosure provides an enterprise evaluation method, the method comprising: for each of a plurality of enterprises to be evaluated, based on different business data information in the tax data of the enterprise to be evaluated, establishing at least two different associated enterprise networks for the enterprise to be evaluated, and determining the weight of the enterprise to be evaluated in each of the associated enterprise networks; wherein the associated enterprise network is used to indicate the relationship information associated with the enterprise to be evaluated and other enterprises;
[0005] According to the weight of the enterprise to be evaluated in each of the associated enterprise networks, adaptive network fusion processing is performed on at least two different associated enterprise networks of the enterprise to be evaluated to obtain data information after network fusion of the enterprise to be evaluated;
[0006] Using the network fused data information of the enterprise to be evaluated through a random walk model, the importance weight of the enterprise to be evaluated is obtained;
[0007] The multiple enterprises to be evaluated are evaluated according to the importance weight of each enterprise to be evaluated.
[0008] Optionally, based on different business data information in the tax data of the enterprise to be assessed, establishing at least two different associated enterprise networks for the enterprise to be assessed includes:
[0009] For each piece of business data information in the tax data of the enterprise to be assessed, a tax meta-network is established according to the information of entities and their corresponding relationships recorded in the business data information;
[0010] If the tax meta-network is a single-entity network, the tax meta-network is determined as an associated enterprise network; if the tax meta-network is not a single-entity network, similar entities are determined from multiple types of entities in the tax meta-network, and corresponding associated enterprise networks are determined based on the similar entities.
[0011] Optionally, the business data information includes: basic enterprise information, value-added tax invoice data information, and goods detailed data information;
[0012] For each piece of business data information in the tax data of the enterprise to be evaluated, establishing a tax meta-network according to the entities and their corresponding relationships recorded in the business data information includes:
[0013] When the business data information is basic enterprise information, an enterprise-natural person meta-network is established based on the relationship information between the enterprise, natural persons, and the enterprise and its corresponding natural persons recorded in the basic enterprise information;
[0014] When the business data information is value-added tax invoice data information, an enterprise-enterprise invoice flow meta-network is established based on the invoice flow transaction relationship information between enterprises and enterprises in the value-added tax invoice data information;
[0015] When the business data information is goods detail data information, an enterprise-purchase and sales goods meta-network is established based on the purchase and sales relationship information between enterprises, goods, and enterprises and goods in the goods detail data information.
[0016] Optionally, if the tax meta network is a single-entity network, determining the tax meta network as an associated enterprise network; if the tax meta network is not a single-entity network, determining similar entities from multiple entities in the tax meta network, and determining corresponding associated enterprise networks according to the similar entities includes:
[0017] When the tax meta-network is an enterprise-natural person meta-network, the enterprise-natural person meta-network is a dual-entity network whose entities are an enterprise and a natural person, the natural person is determined as a similar entity, and the corresponding associated enterprise network is determined based on the natural person;
[0018] When the tax meta-network is an enterprise-enterprise ticket flow meta-network, the enterprise-enterprise ticket flow meta-network is a single entity network, and the enterprise-enterprise ticket flow meta-network is determined to correspond to an associated enterprise network;
[0019] In the enterprise-purchase and sales commodity meta-network, the enterprise-purchase and sales commodity meta-network is a dual-entity network whose entities are the enterprise and the commodity, the commodity is determined as a similar entity, and the corresponding associated enterprise network is determined based on the commodity.
[0020] Optionally, performing adaptive network fusion processing on at least two different associated enterprise networks of the enterprise to be evaluated according to the weight of the enterprise to be evaluated in each of the associated enterprise networks, and obtaining data information after network fusion of the enterprise to be evaluated includes:
[0021] Determine, according to the weight of the enterprise to be evaluated in each of the associated enterprise networks, a transfer matrix corresponding to each of the associated enterprise networks of the enterprise to be evaluated;
[0022] The transfer matrices corresponding to each of the associated enterprise networks of the enterprise to be evaluated are fused according to a preset network weight ratio to obtain a final transfer matrix of the enterprise to be evaluated.
[0023] Optionally, the step of obtaining the importance weight of the enterprise to be evaluated by using the network-fused data information of the enterprise to be evaluated through a random walk model includes:
[0024] The final transfer matrix of the enterprise to be evaluated is used through the random walk module of the PR algorithm to obtain the importance weight of the enterprise to be evaluated.
[0025] Optionally, before establishing at least two different associated enterprise networks for each of the multiple enterprises to be assessed based on different business data information in the tax data of the enterprise to be assessed and determining the weight of the enterprise to be assessed in each of the associated enterprise networks, the method further includes:
[0026] The tax data of each of the plurality of enterprises to be assessed is collected from the tax bureau system.
[0027] Furthermore, the present disclosure also provides an evaluation device for an enterprise, comprising:
[0028] An establishing unit is used to establish at least two different associated enterprise networks for each of the multiple enterprises to be assessed based on different business data information in the tax data of the enterprise to be assessed, and determine the weight of the enterprise to be assessed in each of the associated enterprise networks; wherein the associated enterprise network is used to indicate the relationship information associated with the enterprise to be assessed and other enterprises;
[0029] A processing unit, configured to perform adaptive network fusion processing on at least two different associated enterprise networks of the enterprise to be evaluated according to the weight of the enterprise to be evaluated in each of the associated enterprise networks, and obtain data information after network fusion of the enterprise to be evaluated;
[0030] The calculation unit is further used to obtain the importance weight of the enterprise to be evaluated by using the network fused data information of the enterprise to be evaluated through a random walk model;
[0031] An evaluation unit is used to evaluate the multiple enterprises to be evaluated according to the importance weight of each enterprise to be evaluated.
[0032] Optionally, the establishing unit is specifically used to establish a tax meta-network for each of the different business data information in the tax data of the enterprise to be evaluated, according to the information of the entities and their corresponding relationships recorded in the business data information;
[0033] If the tax meta-network is a single-entity network, the tax meta-network is determined as an associated enterprise network; if the tax meta-network is not a single-entity network, similar entities are determined from multiple types of entities in the tax meta-network, and corresponding associated enterprise networks are determined based on the similar entities.
[0034] Optionally, the business data information includes: basic enterprise information, value-added tax invoice data information, and goods detailed data information;
[0035] The establishment unit is specifically used to establish an enterprise-natural person meta-network according to the relationship information between the enterprise, the natural person, and the enterprise and its corresponding natural person recorded in the basic information of the enterprise when the business data information is the basic information of the enterprise;
[0036] When the business data information is value-added tax invoice data information, an enterprise-enterprise invoice flow meta-network is established based on the invoice flow transaction relationship information between enterprises and enterprises in the value-added tax invoice data information;
[0037] When the business data information is goods detail data information, an enterprise-purchase and sales goods meta-network is established based on the purchase and sales relationship information between enterprises, goods, and enterprises and goods in the goods detail data information.
[0038] Optionally, the establishing unit is specifically configured to, when the tax meta-network is an enterprise-natural person meta-network, the enterprise-natural person meta-network is a dual-entity network in which the entities are an enterprise and a natural person, determine the natural person as a similar entity, and determine a corresponding associated enterprise network based on the natural person;
[0039] When the tax meta-network is an enterprise-enterprise ticket flow meta-network, the enterprise-enterprise ticket flow meta-network is a single entity network, and the enterprise-enterprise ticket flow meta-network is determined to correspond to an associated enterprise network;
[0040] In the enterprise-purchase and sales commodity meta-network, the enterprise-purchase and sales commodity meta-network is a dual-entity network whose entities are the enterprise and the commodity, the commodity is determined as a similar entity, and the corresponding associated enterprise network is determined based on the commodity.
[0041] Optionally, the processing unit is specifically configured to determine a transfer matrix corresponding to each of the associated enterprise networks of the enterprise to be evaluated according to the weight of the enterprise to be evaluated in each of the associated enterprise networks;
[0042] The transfer matrices corresponding to each of the associated enterprise networks of the enterprise to be evaluated are fused according to a preset network weight ratio to obtain a final transfer matrix of the enterprise to be evaluated.
[0043] Optionally, the calculation unit is specifically used to obtain the importance weight of the enterprise to be evaluated by using the final transfer matrix of the enterprise to be evaluated through a random walk module of a PR algorithm.
[0044] Optionally, the method further comprises: a collection unit, configured to collect tax data of each of the plurality of enterprises to be assessed from a tax bureau system.
[0045] Furthermore, the present disclosure also provides a storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the enterprise evaluation method described in the above embodiment are implemented.
[0046] Furthermore, the present disclosure also provides an electronic device, comprising:
[0047] a memory having a computer program stored thereon;
[0048] A processor is used to execute the computer program in the memory to implement the steps of the enterprise evaluation method described in the above embodiment.
[0049] Through the above technical scheme, for each of the multiple enterprises to be evaluated, based on the different business data information in the tax data of the enterprise to be evaluated, at least two different associated enterprise networks are established for the enterprise to be evaluated, and the weight of the enterprise to be evaluated in each associated enterprise network is determined; according to the weight of the enterprise to be evaluated in each associated enterprise network, at least two different associated enterprise networks of the enterprise to be evaluated are adaptively fused to obtain the network fused data information of the enterprise to be evaluated; the importance weight of the enterprise to be evaluated is obtained by using the network fused data information of the enterprise to be evaluated through a random walk model; and multiple enterprises to be evaluated are evaluated according to the importance weight of each enterprise to be evaluated. In this way, by obtaining different business data information in the tax data of the enterprise to be evaluated, multiple different associated enterprise networks can be established, and then the importance weight of the enterprise to be evaluated can be determined according to different associated enterprise networks. In this application, associated enterprise networks can be established from multiple angles, which can improve the accuracy of enterprise importance assessment.
[0050] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:
[0052] Figure 1 It is a flowchart of an enterprise evaluation method provided by this application;
[0053] Figure 2 is an example diagram of a tax meta-network provided by this application;
[0054] Figure 3 is an example diagram of another tax meta-network provided by this application;
[0055] Figure 4 is an example diagram of another tax meta-network provided by this application;
[0056] Figure 5 It is a flowchart of another enterprise evaluation method provided by the present application;
[0057] Figure 6 It is a structural schematic diagram of an evaluation device for an enterprise provided by the present application;
[0058] Figure 7 This is a schematic diagram of the structure of another enterprise evaluation device provided by this application
[0059] Figure 8A schematic diagram of the structure of an electronic device provided by the present application;
[0060] Fig. 9 It is a structural schematic diagram of another electronic device provided by the present application. DETAILED DESCRIPTION
[0061] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.
[0062] In the prior art, the enterprise importance assessment for the tax industry is a technical assessment of a single network data, which has strong limitations and makes the assessed enterprise importance less accurate. In this application, by obtaining different business data information in the tax data of the enterprise to be assessed, multiple different associated enterprise networks can be established, and then the importance weight of the enterprise to be assessed can be determined based on different associated enterprise networks. In this application, associated enterprise networks can be established from multiple angles, which can improve the accuracy of enterprise importance assessment.
[0063] This application provides a method for evaluating an enterprise, which is based on tax data to evaluate the enterprise, such as Figure 1 As shown, the method includes:
[0064] Step S10: for each of the multiple enterprises to be assessed, based on different business data information in the tax data of the enterprise to be assessed, establish at least two different associated enterprise networks for the enterprise to be assessed, and determine the weight of the enterprise to be assessed in each of the associated enterprise networks.
[0065] The associated enterprise network is used to indicate the relationship information between the enterprise to be evaluated and other enterprises.
[0066] It should be noted that other enterprises are enterprises pre-set by the user and associated with the enterprise to be evaluated.
[0067] Specifically, when conducting materiality assessment on multiple enterprises based on their tax data, the enterprises to be assessed may be referred to as enterprises to be assessed. For each of all enterprises to be assessed, the following steps are performed. An enterprise to be assessed is used as an example for explanation.
[0068] Based on different business data information in the tax data of the enterprise to be evaluated, at least two different associated enterprise networks are established for the enterprise to be evaluated, that is, different associated enterprise networks are established according to different business data information in the tax data of the enterprise to be evaluated, wherein one associated enterprise network is established for each business data information. After the associated enterprise networks are established, the weight of the enterprise to be evaluated in each associated enterprise network is determined at the same time.
[0069] Further, based on different business data information in the tax data of the enterprise to be evaluated, establishing at least two different associated enterprise networks for the enterprise to be evaluated includes: for each piece of business data information in the different business data information in the tax data of the enterprise to be evaluated, establishing a tax meta-network according to information on entities and their corresponding relationships recorded in the business data information;
[0070] If the tax meta-network is a single-entity network, the tax meta-network is determined as an associated enterprise network; if the tax meta-network is not a single-entity network, similar entities are determined from multiple entities in the tax meta-network, and corresponding associated enterprise networks are determined based on the similar entities.
[0071] Among them, the tax data contains multiple business data information, such as basic enterprise information, VAT invoice data, purchase and sales information, etc. The business data information records the enterprise to be evaluated and its business data, including at least one type of single entity information of the enterprise to be evaluated and the relationship between the single entity information and the business. Different business data information records different entities and the corresponding relationship between entities. Therefore, tax meta-networks can be established for different business data information, that is, a tax meta-network is established for one business data information, and different tax meta-networks are established for different business data information.
[0072] After multiple tax meta networks are established, if there is a tax meta network containing multiple types of entities among the multiple tax meta networks, it is necessary to convert the tax meta network containing multiple types of entities into a single-entity tax meta network, and this single-entity tax meta network is the associated enterprise network. If there is a single-entity tax meta network among multiple tax meta networks, the single-entity tax meta network is directly determined as the associated enterprise network.
[0073] It should be noted that the tax meta-network of multiple entities is a tax meta-network that simultaneously records multiple categories of entity information of the enterprise to be assessed. For example, if the enterprise and natural person of the enterprise to be assessed are recorded at the same time, then the enterprise and natural person are two types of entities, and this tax meta-network is a multi-entity tax meta-network.
[0074] When converting a tax meta-network containing multiple types of entities into a single-entity tax meta-network, it is necessary to first identify similar entities from the multiple types of entities. The entities of the enterprise to be assessed in the tax meta-network that are the same as those of other enterprises in the tax meta-network can be identified as similar entities. Based on the similar entities, the tax meta-network of multiple types of entities is converted into a single-entity network, and then the single-entity network is identified as a network of associated enterprises.
[0075] Furthermore, the business data information includes: basic enterprise information, value-added tax invoice data information, and goods detailed data information.
[0076] At this time, for each piece of business data information in the tax data of the enterprise to be evaluated, a tax meta-network is established according to the entities and their corresponding relationships recorded in the business data information, including:
[0077] When the business data information is basic enterprise information, an enterprise-natural person meta-network is established based on the relationship information between the enterprise, natural persons, and the enterprise and its corresponding natural persons recorded in the basic enterprise information.
[0078] In this embodiment, the enterprise and natural person entities are first extracted from the basic information of the enterprise, and then the relationship between the enterprise and its corresponding natural person (legal person, financial person in charge, tax payer) is extracted according to each row of records in the basic information data of the taxpayer, and the enterprise-natural person meta-network is constructed, such as Figure 2 shown.
[0079] When the business data information is value-added tax invoice data information, an enterprise-enterprise invoice flow meta-network is established based on the invoice flow transaction relationship information between enterprises and enterprises in the value-added tax invoice data information.
[0080] In this embodiment, the enterprise entities involved in the tax invoice data are first extracted from the value-added tax invoice data information, and then the ticket flow transaction relationship between enterprises is constructed based on the seller-buyer transaction. The relationship attributes are the transaction amount and transaction time between the two enterprise entities, and the enterprise-enterprise ticket flow meta-network is constructed, such as Figure 3 shown.
[0081] When the business data information is goods detail data information, an enterprise-purchase and sales goods meta-network is established based on the purchase and sales relationship information between enterprises, goods, and enterprises and goods in the goods detail data information.
[0082] In this embodiment, the enterprise entities involved in the tax invoice data are first extracted, and the commodity entities are extracted according to the "Commodity and Service Tax Classification Code Table" of the State Administration of Taxation. Then, according to the detailed data information of the goods, the purchase and sales relationship between the enterprise and the purchased and sold commodities is established, and the enterprise-purchase and sold commodity meta-network is constructed. Figure 4 shown.
[0083] At this point, after establishing multiple tax meta-networks, it is necessary to construct an associated enterprise network based on the multiple tax meta-networks. That is, this part is to convert the above-mentioned tax meta-network into an associated enterprise network in which enterprises are directly associated with each other, and convert multiple types of entity associations in a network into single entity associations through similarity relationships.
[0084] If the tax meta-network is a single-entity network, the tax meta-network is determined as an associated enterprise network; if the tax meta-network is not a single-entity network, similar entities are determined from multiple entities in the tax meta-network, and the corresponding associated enterprise networks are determined based on the similar entities, including:
[0085] When the tax meta-network is an enterprise-natural person meta-network, the enterprise-natural person meta-network is a dual-entity network of enterprises and natural persons. Natural persons are identified as similar entities, and the corresponding affiliated enterprise network is determined based on the natural persons.
[0086] In other words, the enterprise-natural person meta-network is a dual-entity network of enterprises and natural persons. The entity natural persons that are the same as the enterprise to be evaluated and other enterprises are regarded as similar entities. From the perspective of the same natural person, a single-entity network based on the similarity between enterprises is constructed, which is the associated enterprise network, denoted by N A The construction method can refer to the following:
[0087] If the company A to be evaluated has the same legal person as the company B: the weight is f;
[0088] If the company A to be evaluated has the same financial director as the company B: the weight is c;
[0089] If the company A to be assessed has the same tax payer as the company B: the weight is b;
[0090] If a natural person of the enterprise A to be evaluated, such as a legal person, is another natural person of other enterprise B, such as a tax payer: the weight is d; among them, f>c>b>d, and is pre-set by the user.
[0091] It should be noted that, in this embodiment, the enterprise to be evaluated is enterprise A, and the other enterprise is other enterprise B.
[0092] After building the network of related enterprises, it is necessary to further determine the weight of the enterprise to be evaluated in this related enterprise. At this time, if the enterprise A to be evaluated has all the above natural person relationships with other enterprises B, in the related enterprise network N AThe weights of the internal bidirectional edges are all f+c+b+d. That is, the weights of the enterprise A to be evaluated and other enterprises B in the network of related enterprises are all f+c+b+d. If there is no natural person relationship between the two enterprises, the two nodes in the network are not connected and the weight is 0. If there is only a partial natural person relationship, that is, the enterprise A to be evaluated and other enterprises B have only some of the same natural persons, then the weight corresponding to the part with the same natural person is determined from the above preset weights (f, c, b, d), and the weights corresponding to the remaining relationships are 0.
[0093] When the tax meta-network is an enterprise-to-enterprise ticket flow meta-network, the enterprise-to-enterprise ticket flow meta-network is a single entity network, and the enterprise-to-enterprise ticket flow meta-network is determined to correspond to an associated enterprise network.
[0094] That is to say, the enterprise-enterprise ticket flow network itself is a single entity network and can be directly used as a network of related enterprises N B , if the enterprise A to be evaluated has a direct ticket flow transaction relationship with other enterprise B, the transaction amount between the enterprise A to be evaluated and the other enterprise B is the weight of each in this associated enterprise network, which are je A→B , je B→A Among them, je A→B Indicates the amount of sales of the company A to be evaluated to other company B, je B→A It indicates the amount sold by other enterprise B to enterprise A to be evaluated.
[0095] In the enterprise-purchase and sales commodity meta-network, the enterprise-purchase and sales commodity meta-network is a dual-entity network whose entities are the enterprise and the commodity, the commodity is determined as a similar entity, and the corresponding associated enterprise network is determined based on the commodity.
[0096] That is to say, the enterprise-purchase and sales commodity meta-network is a dual-entity network with the entity being the enterprise and the commodity. The entity commodities that are the same as those of the enterprise to be evaluated and other enterprises are regarded as similar entities. From the perspective of the similarity of the commodities purchased and sold by the enterprise, a single-entity network based on the similarity between enterprises is constructed, which is the associated enterprise network, denoted by N C .
[0097] At this time, the enterprise A to be evaluated is in this associated enterprise network N C The weight is obtained by the following method: If the set of goods purchased and sold by the company A to be evaluated is C A , the set of goods purchased and sold by other enterprise B is C B , then the enterprise A to be evaluated and the other enterprise B are in the associated enterprise network N C The weights in are:
[0098]
[0099] Among them, N represents the number of overlapping products between the company A to be evaluated and other companies B.
[0100] Step 11: According to the weight of the enterprise to be evaluated in each associated enterprise network, adaptive network fusion processing is performed on at least two different associated enterprise networks of the enterprise to be evaluated to obtain data information after network fusion of the enterprise to be evaluated.
[0101] Specifically, since it is necessary to use the multiple associated enterprise networks established above to evaluate the importance of the enterprise to be evaluated, it is necessary to first merge the multiple associated enterprise networks into one network, and obtain the data information of the enterprise to be evaluated for importance evaluation during the fusion.
[0102] At this time, according to the weight of the enterprise to be evaluated in each of the associated enterprise networks, adaptive network fusion processing is performed on at least two different associated enterprise networks of the enterprise to be evaluated, and data information after network fusion of the enterprise to be evaluated is obtained, including:
[0103] According to the weight of the enterprise to be evaluated in each of the associated enterprise networks, a transfer matrix corresponding to each of the associated enterprise networks of the enterprise to be evaluated is determined.
[0104] The transfer matrices corresponding to each of the associated enterprise networks of the enterprise to be evaluated are fused according to a preset network weight ratio to obtain a final transfer matrix of the enterprise to be evaluated.
[0105] That is to say, for the above-mentioned affiliated enterprise network N A , if the enterprise A to be evaluated and other enterprise B are in the associated enterprise network N A The weight in is f+c+b+d, and the sum of the weights of the edges between the enterprise A to be evaluated and all other enterprises is Weight A , then for the enterprise A to be evaluated, calculate the associated enterprise network N A The corresponding element value of the transfer matrix is (f+c+b+d) / Weight A , at this time, the associated enterprise network N A The corresponding transfer matrix is defined as S1, and the transfer matrix S1 is a diagonal matrix.
[0106] For affiliated enterprise network N B , if the total sales of the company A to be evaluated is Sum A , the amount of sales of enterprise A to be evaluated to other enterprise B A→B , then for the enterprise A to be evaluated, calculate the associated enterprise network N B The corresponding element value of the transfer matrix is je A→B / Sum A , the transfer matrix is defined as S2.
[0107] For affiliated enterprise network N C , if the set of goods purchased and sold by the company A to be evaluated is C A , the set of goods purchased and sold by other enterprise B is C B , in the affiliated enterprise network N C The weights of the enterprise A to be evaluated and other enterprises B are calculated by formula 1. The weights are normalized by column to obtain the associated enterprise network N. C The corresponding transfer matrix S3.
[0108] In the context of the affiliated enterprise network N A , N B , N C After establishing the corresponding transfer matrices, the three corresponding transfer matrices need to be fused. At this time, the three corresponding transfer matrices need to be fused according to the preset network weight ratio. The adaptive network fusion processing of at least two different associated enterprise networks of the enterprise to be evaluated is actually to merge the three associated enterprise networks N A 、N B 、N C , merged into a network through a predefined weight ratio, here is the associated enterprise network N A The weight assigned is n1, which is the associated enterprise network N B The weight assigned is n2, which is the associated enterprise network N C The weight is assigned to n3, and the three corresponding transfer matrices are fused to obtain the final transfer matrix S of the enterprise A to be evaluated.
[0109] Specifically, the final transfer matrix S can be obtained by the following formula.
[0110]
[0111] The sum of the element values in each column of S is 1.
[0112] It should be noted that n1, n2, and n3 are preset by the user according to actual needs.
[0113] The requirement is that the business data information based on which the enterprise to be evaluated needs to establish an associated enterprise network is pre-set by the user according to actual needs. When setting how many associated enterprise networks need to be established, the weight ratio of the transfer matrix corresponding to each associated enterprise network when performing network fusion can be set at the same time.
[0114] Step S12: using the network-fused data information of the enterprise to be evaluated through a random walk model to obtain the importance weight of the enterprise to be evaluated.
[0115] Specifically, after calculating the network-fused data information of the enterprise to be evaluated, the data information is processed using a random walk model to obtain the importance weight of the enterprise to be evaluated.
[0116] Furthermore, the random walk module may be a PR algorithm random walk module. In this case, the importance weight of the enterprise to be evaluated is obtained by using the network fused data information of the enterprise to be evaluated through a random walk model, including:
[0117] The final transfer matrix of the enterprise to be evaluated is used through the random walk module of the PR (PageRank) algorithm to obtain the importance weight of the enterprise to be evaluated.
[0118] Specifically, according to the final transfer matrix S after fusion, the initialization transfer matrix S' of the PR algorithm is defined, where S' is calculated by the following formula three.
[0119]
[0120] Among them, α is a constant coefficient, α∈[0,1), generally α=0.85, and the power iteration method is used to converge the initialized transfer matrix S' to a stable vector to obtain the importance weight.
[0121] It should be noted that the PR algorithm is: assign an initial value P1 to each node. For example, the initial column vector P1 can be [1 / n, 1 / n, 1 / n, ...], where n is an integer. The PR value is calculated as follows, where P n is the column vector composed of the PR values of each node at the nth iteration:
[0122] P n+1 =S'P n Formula 4
[0123] The PR value is continuously iterated through Formula 4. When the inequality of Formula 5 is satisfied, the iteration ends and the PR value of all nodes is obtained. ε is the error bound:
[0124] |P n+1 -P n |<ε Formula 5
[0125] It should be noted that ε is preset by the user.
[0126] In this way, through the random walk model of the above PR algorithm, the PR value at the end of the last iteration can be used as the importance weight of the enterprise to be evaluated.
[0127] Step S13: Evaluate multiple enterprises to be evaluated according to the importance weight of each enterprise to be evaluated.
[0128] Specifically, after obtaining the importance weights of all the companies to be evaluated, the importance weights of all the companies to be evaluated can be sorted, for example, sorted from large to small, to achieve the ranking of the importance of the companies to be evaluated, and then the companies to be focused on can be determined based on this ranking.
[0129] like Figure 5 As shown, in the present application, before the above step S10, it also includes:
[0130] Step S14: Collect tax data of each of the multiple enterprises to be assessed from the tax bureau system.
[0131] Specifically, since the assessment of enterprises is based on tax data, the tax data of all enterprises to be assessed can be collected in the tax bureau system.
[0132] In this way, by obtaining different business data information in the tax data of the enterprise to be evaluated, multiple different associated enterprise networks can be established, and then the importance weight of the enterprise to be evaluated can be determined based on different associated enterprise networks. In this application, the associated enterprise network can be established from multiple angles, which can improve the accuracy of the enterprise importance assessment.
[0133] like Figure 6 As shown, the present disclosure also provides an enterprise evaluation device 600, comprising:
[0134] The establishing unit 601 is used to establish at least two different associated enterprise networks for each of the multiple enterprises to be assessed based on different business data information in the tax data of the enterprise to be assessed, and determine the weight of the enterprise to be assessed in each of the associated enterprise networks.
[0135] The associated enterprise network is used to indicate the relationship information between the enterprise to be evaluated and other enterprises.
[0136] The processing unit 602 is used to perform adaptive network fusion processing on at least two different associated enterprise networks of the enterprise to be evaluated according to the weight of the enterprise to be evaluated in each of the associated enterprise networks, and obtain data information of the enterprise to be evaluated after network fusion.
[0137] The calculation unit 603 is further used to obtain the importance weight of the enterprise to be evaluated by using the network fused data information of the enterprise to be evaluated through a random walk model.
[0138] The evaluation unit 604 is used to evaluate the multiple enterprises to be evaluated according to the importance weight of each enterprise to be evaluated.
[0139] Optionally, the establishing unit 601 is specifically configured to establish a tax meta-network for each of the different business data information in the tax data of the enterprise to be evaluated, according to the information of entities and their corresponding relationships recorded in the business data information;
[0140] If the tax meta-network is a single-entity network, the tax meta-network is determined as an associated enterprise network; if the tax meta-network is not a single-entity network, similar entities are determined from multiple types of entities in the tax meta-network, and corresponding associated enterprise networks are determined based on the similar entities.
[0141] Optionally, the business data information includes: basic enterprise information, value-added tax invoice data information, and goods detailed data information;
[0142] The establishment unit 601 is specifically used to establish an enterprise-natural person meta-network according to the relationship information between the enterprise, the natural person, and the enterprise and its corresponding natural person recorded in the basic information of the enterprise when the business data information is the basic information of the enterprise;
[0143] When the business data information is value-added tax invoice data information, an enterprise-enterprise invoice flow meta-network is established based on the invoice flow transaction relationship information between enterprises and enterprises in the value-added tax invoice data information;
[0144] When the business data information is goods detail data information, an enterprise-purchase and sales goods meta-network is established based on the purchase and sales relationship information between enterprises, goods, and enterprises and goods in the goods detail data information.
[0145] Optionally, the establishing unit 601 is specifically configured to, when the tax meta-network is an enterprise-natural person meta-network, the enterprise-natural person meta-network is a dual-entity network in which the entities are an enterprise and a natural person, determine the natural person as a similar entity, and determine a corresponding associated enterprise network according to the natural person;
[0146] When the tax meta-network is an enterprise-enterprise ticket flow meta-network, the enterprise-enterprise ticket flow meta-network is a single entity network, and the enterprise-enterprise ticket flow meta-network is determined to correspond to an associated enterprise network;
[0147] In the enterprise-purchase and sales commodity meta-network, the enterprise-purchase and sales commodity meta-network is a dual-entity network whose entities are the enterprise and the commodity, the commodity is determined as a similar entity, and the corresponding associated enterprise network is determined based on the commodity.
[0148] Optionally, the processing unit 602 is specifically configured to determine a transfer matrix corresponding to each of the associated enterprise networks of the enterprise to be evaluated according to the weight of the enterprise to be evaluated in each of the associated enterprise networks;
[0149] The transfer matrices corresponding to each of the associated enterprise networks of the enterprise to be evaluated are fused according to a preset network weight ratio to obtain a final transfer matrix of the enterprise to be evaluated.
[0150] Optionally, the calculation unit 603 is specifically configured to obtain the importance weight of the enterprise to be evaluated by using the final transfer matrix of the enterprise to be evaluated through a random walk module of a PR algorithm.
[0151] Alternatively, if Figure 7 As shown, the enterprise evaluation device 600 further includes: a collection unit 604, which is used to collect tax data of each of the multiple enterprises to be evaluated from the tax bureau system.
[0152] In this way, by obtaining different business data information in the tax data of the enterprise to be evaluated, multiple different associated enterprise networks can be established, and then the importance weight of the enterprise to be evaluated can be determined based on different associated enterprise networks. In this application, the associated enterprise network can be established from multiple angles, which can improve the accuracy of the enterprise importance assessment.
[0153] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.
[0154] Furthermore, the present disclosure also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the enterprise evaluation method described in the above embodiment are implemented.
[0155] Figure 8 FIG. 7 is a block diagram of an electronic device 700 according to an exemplary embodiment. Figure 8 As shown, the electronic device 700 may include: a processor 701 and a memory 702. The electronic device 700 may also include one or more of a multimedia component 703, an input / output (I / O) interface 704, and a communication component 705.
[0156] The processor 701 is used to control the overall operation of the electronic device 700 to complete all or part of the steps in the above-mentioned enterprise evaluation method. The memory 702 is used to store various types of data to support the operation of the electronic device 700, and these data may include, for example, instructions for any application or method used to operate on the electronic device 700, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 703 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 702 or sent through the communication component 705. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 704 provides an interface between the processor 701 and other interface modules, and the other interface modules may be keyboards, mice, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more of them, is not limited here. Therefore, the corresponding communication component 705 may include: Wi-Fi module, Bluetooth module, NFC module, etc.
[0157] In an exemplary embodiment, the electronic device 700 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned enterprise evaluation method.
[0158] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned enterprise evaluation method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 702 including program instructions, and the above-mentioned program instructions can be executed by the processor 701 of the electronic device 700 to complete the above-mentioned enterprise evaluation method.
[0159] Fig. 9 1 is a block diagram of an electronic device 1900 according to an exemplary embodiment. For example, the electronic device 1900 may be provided as a server. Fig. 9 The electronic device 1900 includes a processor 1922, which may be one or more, and a memory 1932 for storing a computer program executable by the processor 1922. The computer program stored in the memory 1932 may include one or more modules each corresponding to a set of instructions. In addition, the processor 1922 may be configured to execute the computer program to perform the above-mentioned enterprise evaluation method.
[0160] In addition, the electronic device 1900 may further include a power supply component 1926 and a communication component 1950. The power supply component 1926 may be configured to perform power management of the electronic device 1900, and the communication component 1950 may be configured to implement communication of the electronic device 1900, for example, wired or wireless communication. In addition, the electronic device 1900 may further include an input / output (I / O) interface 1958. The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server 2003. TM , Mac OS X TM , Unix TM , Linux TM etc.
[0161] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned enterprise evaluation method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 1932 including program instructions, and the above-mentioned program instructions can be executed by the processor 1922 of the electronic device 1900 to complete the above-mentioned enterprise evaluation method.
[0162] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program executable by a programmable device, and the computer program has a code portion for executing the above-mentioned enterprise evaluation method when executed by the programmable device.
[0163] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.
[0164] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.
[0165] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.
Claims
1. A method for evaluating an enterprise, characterized in that: The method comprises: For each of the multiple enterprises to be assessed, based on different business data information in the tax data of the enterprise to be assessed, at least two different associated enterprise networks are established for the enterprise to be assessed, and the weight of the enterprise to be assessed in each of the associated enterprise networks is determined; wherein the associated enterprise network is used to indicate the relationship information between the enterprise to be assessed and other enterprises; different business data information is used to establish different associated enterprise networks; the business data information includes: basic enterprise information, value-added tax invoice data information, and goods detailed data information; According to the weight of the enterprise to be evaluated in each of the associated enterprise networks, performing adaptive network fusion processing on at least two different associated enterprise networks of the enterprise to be evaluated to obtain data information after network fusion of the enterprise to be evaluated; The importance weight of the enterprise to be evaluated is obtained by using the network fusion data information of the enterprise to be evaluated through a random walk model; Evaluate the multiple enterprises to be evaluated according to the importance weight of each enterprise to be evaluated; Based on different business data information in the tax data of the enterprise to be assessed, establishing at least two different associated enterprise networks for the enterprise to be assessed includes: For each piece of business data information in the tax data of the enterprise to be assessed, a tax meta-network is established based on the entities and their corresponding relationships recorded in the business data information; the tax meta-network includes any one of an enterprise-natural person meta-network established based on the basic information of the enterprise, an enterprise-enterprise ticket flow meta-network established based on the value-added tax invoice data information, and an enterprise-purchase and sale commodity meta-network established based on the goods detail data information; If the tax meta-network is a single-entity network, the tax meta-network is determined as an associated enterprise network; if the tax meta-network is not a single-entity network, similar entities are determined from multiple types of entities in the tax meta-network, and corresponding associated enterprise networks are determined based on the similar entities; The similar entity is an entity of the enterprise to be assessed that is the same as other enterprises in the tax meta-network.
2. The method according to claim 1, characterized in that For each piece of business data information in the tax data of the enterprise to be evaluated, establishing a tax meta-network according to the entities and their corresponding relationships recorded in the business data information includes: When the business data information is basic enterprise information, an enterprise-natural person meta-network is established based on the relationship information between the enterprise, natural persons, and the enterprise and its corresponding natural persons recorded in the basic enterprise information; When the business data information is value-added tax invoice data information, an enterprise-enterprise invoice flow meta-network is established based on the invoice flow transaction relationship information between enterprises and enterprises in the value-added tax invoice data information; When the business data information is goods detail data information, an enterprise-purchase and sales goods meta-network is established based on the purchase and sales relationship information between enterprises, goods, and enterprises and goods in the goods detail data information.
3. The method according to claim 2, characterized in that If the tax meta-network is a single-entity network, the tax meta-network is determined as an associated enterprise network; if the tax meta-network is not a single-entity network, similar entities are determined from multiple entities in the tax meta-network, and the corresponding associated enterprise network is determined based on the similar entities, including: When the tax meta-network is an enterprise-natural person meta-network, the enterprise-natural person meta-network is a dual-entity network whose entities are an enterprise and a natural person, the natural person is determined as a similar entity, and the corresponding associated enterprise network is determined based on the natural person; When the tax meta-network is an enterprise-enterprise ticket flow meta-network, the enterprise-enterprise ticket flow meta-network is a single entity network, and the enterprise-enterprise ticket flow meta-network is determined to correspond to an associated enterprise network; In the enterprise-purchase and sales commodity meta-network, the enterprise-purchase and sales commodity meta-network is a dual-entity network whose entities are the enterprise and the commodity, the commodity is determined as a similar entity, and the corresponding associated enterprise network is determined based on the commodity.
4. The method according to claim 1, characterized in that: The step of performing adaptive network fusion processing on at least two different associated enterprise networks of the enterprise to be evaluated according to the weight of the enterprise to be evaluated in each of the associated enterprise networks, and obtaining data information after network fusion of the enterprise to be evaluated comprises: Determine, according to the weight of the enterprise to be evaluated in each of the associated enterprise networks, a transfer matrix corresponding to each of the associated enterprise networks of the enterprise to be evaluated; The transfer matrices corresponding to each of the associated enterprise networks of the enterprise to be evaluated are fused according to a preset network weight ratio to obtain a final transfer matrix of the enterprise to be evaluated.
5. The method according to claim 4, characterized in that: The method of obtaining the importance weight of the enterprise to be evaluated by using the network fused data information of the enterprise to be evaluated through a random walk model includes: The final transfer matrix of the enterprise to be evaluated is used through the random walk module of the PR algorithm to obtain the importance weight of the enterprise to be evaluated.
6. The method according to claim 1, characterized in that Before establishing at least two different associated enterprise networks for each of the multiple enterprises to be assessed based on different business data information in the tax data of the enterprise to be assessed and determining the weight of the enterprise to be assessed in each of the associated enterprise networks, the method further includes: The tax data of each of the plurality of enterprises to be assessed is collected from the tax bureau system.
7. An evaluation device for an enterprise, characterized in that: include: An establishing unit is used to establish at least two different associated enterprise networks for each of the multiple enterprises to be assessed based on different business data information in the tax data of the enterprise to be assessed, and determine the weight of the enterprise to be assessed in each of the associated enterprise networks; wherein the associated enterprise network is used to indicate the relationship information between the enterprise to be assessed and other enterprises; different business data information is used to establish different associated enterprise networks; the business data information includes: basic enterprise information, value-added tax invoice data information, and goods detailed data information; A processing unit, configured to perform adaptive network fusion processing on at least two different associated enterprise networks of the enterprise to be evaluated according to the weight of the enterprise to be evaluated in each of the associated enterprise networks, and obtain data information after network fusion of the enterprise to be evaluated; The calculation unit is further used to obtain the importance weight of the enterprise to be evaluated by using the network fused data information of the enterprise to be evaluated through a random walk model; An evaluation unit, configured to evaluate the plurality of enterprises to be evaluated according to the importance weight of each enterprise to be evaluated; The establishment unit is used to establish a tax meta-network for each of the different business data information in the tax data of the enterprise to be evaluated, according to the information of the entities and their corresponding relationships recorded in the business data information; the tax meta-network includes any one of an enterprise-natural person meta-network established according to the basic information of the enterprise, an enterprise-enterprise ticket flow meta-network established according to the value-added tax invoice data information, and an enterprise-purchase and sale commodity meta-network established according to the goods detail data information; if the tax meta-network is a single-entity network, the tax meta-network is determined as an associated enterprise network; if the tax meta-network is not a single-entity network, similar entities are determined from multiple types of entities in the tax meta-network, and corresponding associated enterprise networks are determined based on the similar entities; wherein the similar entities are the same entities in the tax meta-network of the enterprise to be evaluated and other enterprises in the tax meta-network.
8. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.
9. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 6.
