Intelligent tax planning methods and related equipment

Abstract

The embodiment of the present application discloses a tax intelligent planning method and related equipment for improving the reliability of recommended tax schemes. The method of the embodiment of the present application includes: obtaining an available rule set corresponding to the enterprise to be planned; determining the dimension value of each available tax calculation rule in each tax calculation dimension, and based on the dimension value of each available tax calculation rule in each tax calculation dimension and the objective weight calculation method, calculating the objective weight of each tax calculation rule in each tax calculation dimension; determining each tax calculation dimension of each available tax calculation rule as a target node, and determining the target weight corresponding to each target node and the product of the dimension value corresponding to each target node as the node value of each target node; taking each target node as a path node, the node value of each target node as the path length, and determining a group of target nodes as a recommended node set through a path optimization algorithm; determining a recommended tax scheme based on the target available tax calculation rule and the target tax calculation dimension corresponding to each target node.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of tax software, and in particular to intelligent tax planning methods and related equipment. Background Art

[0002] Tax planning refers to a planning method that aims to obtain the maximum possible tax benefits through advance planning and arrangement of business, investment, financial management and other activities within the scope prescribed by tax laws.

[0003] In existing solutions, users typically provide a set of available tax rules for the business to be planned. This set contains every applicable tax calculation rule that the business can use. A brute force calculation is then performed to determine the total tax burden corresponding to different combinations of available tax calculation rules. Ultimately, the combination with the lowest total tax burden is selected as the recommended tax solution.

[0004] The recommended tax plan calculated by this brute force calculation method is not necessarily the most suitable tax plan for the enterprise to be planned and the one with the lowest overall tax burden, and its reliability is low. Summary of the Invention

[0005] The embodiments of the present application provide a tax intelligent planning method and related equipment for improving the reliability of recommended tax plans.

[0006] A first aspect of the embodiments of the present application provides a method for intelligent tax planning, comprising:

[0007] Obtaining an available rule set corresponding to the enterprise to be planned, wherein the available rule set corresponds to multiple available tax calculation rules;

[0008] Determining a dimension value of each available tax calculation rule in each tax calculation dimension of the available rule set, and calculating an objective weight of each tax calculation rule in each tax calculation dimension based on the dimension value of each available tax calculation rule in each tax calculation dimension and an objective weight calculation method;

[0009] Determine each tax calculation dimension of each available tax calculation rule as a target node, and determine the product of a target weight corresponding to each target node and a dimension value corresponding to each target node as the node value of each target node; the target weight corresponding to each target node includes the objective weight of the available tax calculation rule corresponding to each target node under the corresponding tax calculation dimension;

[0010] Taking each target node as a path node and the node value of each target node as the path length, a group of target nodes whose total path length meets a preset recommendation condition is determined as a recommended node set by a path optimization algorithm;

[0011] A recommended tax plan is determined based on the target available tax calculation rules and target tax calculation dimensions corresponding to each target node in the recommended node set.

[0012] In a specific implementation, the target weight corresponding to each target node further includes a subjective weight of the available tax calculation rule corresponding to each target node under the corresponding tax calculation dimension, and the method further includes:

[0013] Obtain any historical recommended tax plan and the corresponding historical actual tax plan;

[0014] Calculate the deviation value of each applicable tax calculation rule included in the historical recommended tax plan and the historical actual tax plan under each tax calculation dimension;

[0015] Based on the deviation value of each available tax calculation rule in each tax calculation dimension and the objective weight calculation method, the subjective weight of each available tax calculation rule in each tax calculation dimension is calculated.

[0016] In a specific implementation, the group of target nodes whose total path lengths satisfy preset recommendation conditions determined by the path optimization algorithm is the recommended node set, including:

[0017] A set of target nodes with the longest sum of corresponding path lengths is determined by a path optimization algorithm, and is referred to as the recommended node set.

[0018] In a specific implementation, the method further includes:

[0019] Inputting the enterprise information of the enterprise to be planned into a pre-built portrait decision tree to determine the enterprise portrait;

[0020] The enterprise portrait is input into a pre-built rule decision tree to determine the available rule set, which includes multiple available tax calculation rules corresponding to the enterprise portrait, and each tax calculation rule includes at least one tax calculation dimension.

[0021] In a specific implementation, the method further includes:

[0022] Constructing the portrait decision tree based on multiple types of enterprise policies in the enterprise policy database;

[0023] The rule decision tree is constructed based on multiple tax calculation rules in a tax policy database.

[0024] In a specific implementation, inputting the enterprise profile into a pre-built rule decision tree to determine the available rule set includes:

[0025] Obtain at least one rule selection constraint input by a user;

[0026] The at least one rule selection constraint and the enterprise profile are input into the rule decision tree to obtain the available rule set.

[0027] In a specific implementation, the available rule set further includes a mutually exclusive relationship between any two available tax calculation rules; and there is no mutually exclusive relationship between the two available tax calculation rules corresponding to any two target nodes in the recommended node set.

[0028] A second aspect of the present application provides a tax intelligent planning device, comprising:

[0029] An acquisition unit, configured to acquire an available rule set corresponding to the enterprise to be planned, wherein the available rule set corresponds to a plurality of available tax calculation rules;

[0030] a weight calculation unit, configured to determine a dimension value of each available tax calculation rule in the available rule set in each tax calculation dimension, and calculate an objective weight of each tax calculation rule in each tax calculation dimension based on the dimension value of each available tax calculation rule in each tax calculation dimension and an objective weight calculation method;

[0031] a determination unit, configured to determine each tax calculation dimension of each available tax calculation rule as a target node, and determine the product of a target weight corresponding to each target node and a dimension value corresponding to each target node as the node value of each target node; the target weight corresponding to each target node includes an objective weight of the available tax calculation rule corresponding to each target node under the corresponding tax calculation dimension;

[0032] The determining unit is further configured to use each target node as a path node and the node value of each target node as the path length, and determine, through a path optimization algorithm, a group of target nodes whose total path length meets a preset recommendation condition as a recommended node set;

[0033] The determining unit is further configured to determine a recommended tax plan based on the target available tax calculation rules and target tax calculation dimensions corresponding to each target node in the recommended node set.

[0034] In a specific implementation, the target weight corresponding to each target node further includes the subjective weight of the available tax calculation rules corresponding to each target node under the corresponding tax calculation dimension, and the device further includes:

[0035] The acquisition unit is further configured to acquire any historically recommended tax plan and the corresponding historically actual tax plan;

[0036] The weight calculation unit is further configured to calculate a deviation value between each applicable tax calculation rule included in the historical recommended tax plan and the historical actual tax plan under each tax calculation dimension;

[0037] The weight calculation unit is further used to calculate the subjective weight of each available tax calculation rule in each tax calculation dimension based on the deviation value of each available tax calculation rule in each tax calculation dimension and the objective weight calculation method.

[0038] In a specific implementation, the determining unit is specifically configured to determine, through a path optimization algorithm, a group of target nodes with the longest sum of corresponding path lengths, as the recommended node set.

[0039] In a specific implementation, the determining unit is further configured to input the enterprise information of the enterprise to be planned into a pre-built portrait decision tree to determine the enterprise portrait;

[0040] The determination unit is also used to input the enterprise portrait into a pre-built rule decision tree to determine the available rule set, which includes multiple available tax calculation rules corresponding to the enterprise portrait, and each tax calculation rule includes at least one tax calculation dimension.

[0041] In a specific implementation, the apparatus further includes: a construction unit;

[0042] The construction unit is used to construct the portrait decision tree based on multiple types of enterprise policies in the enterprise policy database;

[0043] The construction unit is further used to construct the rule decision tree based on multiple tax calculation rules in the tax policy database.

[0044] In a specific implementation, the determining unit is specifically configured to obtain at least one rule selection constraint input by a user;

[0045] The at least one rule selection constraint and the enterprise profile are input into the rule decision tree to obtain the available rule set.

[0046] In a specific implementation, the available rule set further includes a mutually exclusive relationship between any two available tax calculation rules; and there is no mutually exclusive relationship between the two available tax calculation rules corresponding to any two target nodes in the recommended node set.

[0047] A third aspect of the present application provides a tax intelligent planning device, comprising:

[0048] CPU, memory and input / output interfaces;

[0049] The memory is a transient storage memory or a persistent storage memory;

[0050] The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the method described in the first aspect.

[0051] A fourth aspect of the embodiments of the present application provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the method described in the first aspect.

[0052] A fifth aspect of an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores instructions. When the instructions are executed on a computer, the computer executes the method described in the first aspect.

[0053] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: Based on the objective weight calculation method and the dimension value of each applicable tax calculation rule in each tax calculation dimension, the objective weight of each tax calculation rule is determined. Furthermore, each target node is cleverly used as a path node, and the node value of each target node is used as the path length. A recommended tax plan is ultimately obtained through a path optimization algorithm. By taking into account the objective weight of each tax calculation rule, the calculated recommended tax plan is more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A system architecture diagram of the tax intelligent planning system disclosed in the embodiment of this application;

[0055] Figure 2 A schematic diagram of a process flow of the intelligent tax planning method disclosed in an embodiment of the present application;

[0056] Figure 3 This is another flowchart of the tax intelligent planning method disclosed in the embodiment of this application;

[0057] Figure 4 A flowchart of a method for obtaining an available rule set disclosed in an embodiment of the present application;

[0058] Figure 5 This is a structural diagram of the intelligent tax planning device disclosed in the embodiment of this application;

[0059] Figure 6 This is another structural diagram of the intelligent tax planning device disclosed in an embodiment of this application. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0061] The embodiments of the present application provide a tax intelligent planning method and related equipment for improving the reliability of recommended tax plans.

[0062] See also Figure 1 To better implement intelligent tax planning methods, the present application provides an intelligent tax planning system, comprising an intelligent tax planning device 101. The intelligent tax planning device 101 can determine the available tax calculation rules, i.e., an available rule set, for an enterprise requiring tax planning (i.e., the enterprise to be planned). The device 101 then determines the dimension value of each available tax calculation rule in the available rule set for each tax calculation dimension. Next, based on the dimension value of each available tax calculation rule and an objective weight calculation method, the objective weight of each available tax calculation rule is determined. Furthermore, using each tax calculation dimension of each available rule as a target node and the product of the target weight and the dimension value corresponding to each target node as the path length, a path optimization algorithm is used to determine a set of target nodes whose total path length meets preset recommendation criteria, thereby forming a recommended node set. Finally, the target available tax calculation rule and the corresponding target tax calculation dimension corresponding to each target node in the recommended node set are used as a recommended tax plan. In other words, each target available tax calculation rule and the target tax calculation dimension corresponding to each target available tax calculation rule included in the recommended tax plan correspond to a target node in the recommended node set.

[0063] See below Figure 2 Based on the aforementioned intelligent tax planning system, the present embodiment provides an intelligent tax planning method. The method can be executed by the aforementioned intelligent tax planning device 101. The method includes the following steps:

[0064] 201. Obtain the available rule set corresponding to the enterprise to be planned. The available rule set corresponds to multiple available tax calculation rules.

[0065] When it is necessary to recommend a tax plan to a company to be planned, the first step is to identify every available tax calculation rule that the company can use from the massive amount of tax calculation rules, that is, to obtain the available rule set corresponding to the company to be planned.

[0066] 202. Determine the dimension value of each available tax calculation rule in each tax calculation dimension in the available rule set, and calculate the objective weight of each tax calculation rule in each tax calculation dimension based on the dimension value of each available tax calculation rule in each tax calculation dimension and the objective weight calculation method.

[0067] Based on the available rule set obtained in step 201, determine the dimension value of each available tax calculation rule in different tax calculation dimensions. It should be noted that the tax calculation dimensions may include but are not limited to tax rates and / or tax exemptions, etc. Then, the objective weight of each tax calculation rule in each tax calculation dimension is calculated by the objective weight calculation method and the dimension value. Among them, the objective weight calculation method includes but is not limited to: entropy weight method, factor analysis method, principal component method, AHP hierarchical method and priority diagram method, etc. In addition, the calculation method of objective weight can refer to Figure 3 The relevant process of objective weight calculation in .

[0068] 203. Determine each tax calculation dimension of each available tax calculation rule as a target node, and determine the target weight corresponding to each target node and the product of the dimension value corresponding to each target node as the node value of each target node; the target weight corresponding to each target node includes the objective weight of the available tax calculation rule corresponding to each target node under the corresponding tax calculation dimension.

[0069] Drawing on the principles of the Dijkstra algorithm, each tax calculation dimension of each applicable tax calculation rule is treated as a target node, and the node value of each target node is the product of the target weight corresponding to each target node and the corresponding dimension value. In other words, if there are j applicable tax calculation rules and i tax calculation dimensions, there are i × j target nodes. It should be noted that the target weight corresponding to a target node includes the objective weight of the applicable tax calculation rule corresponding to the target node within the tax calculation dimension corresponding to the target node.

[0070] 204. Taking each target node as a path node and the node value of each target node as the path length, a group of target nodes whose total path length meets a preset recommendation condition is determined as a recommended node set through a path optimization algorithm.

[0071] Based on step 203, each target node is used as a path node, and the node value of each target node is used as the path length. A path optimization algorithm is used to determine a group of target nodes whose total path length meets the preset recommendation criteria. These target nodes are then referred to as the recommended node set. Path optimization algorithms include, but are not limited to, Dijkstra's algorithm, dynamic programming algorithms, and brute force enumeration algorithms.

[0072] Specifically, each target node corresponds to an available tax calculation rule and any tax calculation dimension of that available tax calculation rule, so each set of target nodes has a corresponding tax solution. All that needs to be done is to use the path optimization algorithm to select a set of target nodes that meet the preset recommendation criteria as the recommended node set.

[0073] In practical applications, the set of target nodes with the longest total path length can be selected from each possible set of target nodes as the recommended node set. It should be understood that the total path length of each set of target nodes reflects the overall tax burden of the tax plan corresponding to that set of target nodes. The longer the total path length, the lower the overall tax burden.

[0074] 205. Determine a recommended tax plan based on the target available tax calculation rules and target tax calculation dimensions corresponding to each target node in the recommended node set.

[0075] According to step 204, the recommended node set corresponds to a tax plan, namely, the recommended tax plan. Specifically, the target available tax calculation rule and the target tax calculation dimension corresponding to each target node in the recommended node set are used as part of the recommended tax plan, ultimately obtaining a recommended tax plan that includes each target available tax calculation rule and the target tax calculation dimension corresponding to each target available tax calculation rule.

[0076] In this embodiment, the objective weight of each applicable tax calculation rule is determined based on an objective weight calculation method and the value of each applicable tax calculation rule in each tax calculation dimension. A path optimization algorithm cleverly uses each target node as a path node and the node value of each target node as the path length to ultimately determine a recommended tax plan. By considering the objective weight of each tax calculation rule, the resulting recommended tax plan is more reliable.

[0077] See also Figure 3 On the basis of the above-mentioned embodiment, the target weight of the target node in the above-mentioned step 203 also includes the available tax calculation rules corresponding to the target node and the subjective weight under the tax calculation dimension corresponding to the target node. The embodiment of the present application also includes: obtaining any historical recommended tax plan (any recommended tax plan) and the corresponding historical actual tax plan (the corresponding actual tax plan); calculating the deviation value of each available tax calculation rule contained in the historical recommended tax plan and the historical actual tax plan under each tax calculation dimension; based on the deviation value of each available tax calculation rule under each tax calculation dimension and the objective weight calculation method, calculating the subjective weight of each available tax calculation rule under each tax calculation dimension.

[0078] In other words, the subjective weight of each available tax calculation rule in each tax calculation dimension can be obtained by studying the deviation between the historical recommended tax scheme and the corresponding historical actual tax scheme. Specifically, the deviation value of each available tax calculation rule contained in the historical recommended tax scheme and the historical actual tax scheme in each tax calculation dimension is calculated. Then, referring to the aforementioned method of calculating the objective weight of each available tax calculation rule in each tax calculation dimension based on the dimension value, the subjective weight of each available tax calculation rule in each tax calculation dimension is calculated based on the deviation value. It should be noted that although the objective weight calculation method is adopted, the deviation value (rather than the dimension value) is used as the calculation object here, taking into account the deviation between the user's choice (historical actual tax scheme) and the recommendation (historical recommended tax scheme), which is equivalent to the subjective weight.

[0079] In addition, the above deviation value calculation formula can refer to the following formula:

[0080] A ij =F ij -X ij

[0081] Among them, A ij is the deviation value of the jth available tax calculation rule under the i-th tax calculation dimension, F ij X is the dimension value of the jth available tax calculation rule under the i-th tax calculation dimension in the historical actual tax plan, ij It is the dimension value of the jth available tax calculation rule under the i-th tax calculation dimension in the historically recommended tax plan.

[0082] It should be noted that in actual applications, each time intelligent tax planning is conducted, the available tax calculation rules corresponding to the enterprise to be planned will be different. However, when learning active weights, each available tax calculation rule included in the historical recommended tax plans and historical actual tax plans can be learned. Therefore, when determining the target weight, if the available tax calculation rule has a corresponding subjective weight under a certain tax calculation dimension, then the target weight of the target node corresponding to the tax calculation rule and the aforementioned tax calculation dimension is the sum of the corresponding objective weight and the corresponding subjective weight; if the available tax calculation rule does not have a corresponding subjective weight under a certain tax calculation dimension, then the target weight of the target node corresponding to the tax calculation rule and the aforementioned tax calculation dimension only includes the objective weight.

[0083] In an embodiment of the present application, taking into account the needs of the user, based on the deviation between the historical actual tax plan and the historical recommended tax plan, the user's subjective weight for each available tax calculation rule included in both the historical actual tax plan and the historical recommended tax plan under each tax calculation dimension is learned. At the same time, for the target node with a corresponding subjective weight, the target weight of the target node includes the corresponding objective weight and the corresponding subjective weight. Self-learning to obtain subjective weights realizes the technical effect of reinforcement learning, so that the recommended tax plan obtained can better meet the needs of the user (or the enterprise to be planned) and realize the true maximization of corporate interests.

[0084] In practice, tax calculation rules apply to numerous items, which change over time. The criteria and dimensions for judgment are extremely complex, encompassing, but not limited to, taxpayer qualifications (e.g., export tax rebates require a distinction between foreign trade enterprises and manufacturing enterprises; VAT micro-tax exemptions require a distinction between small-scale and general taxpayers), region (e.g., the Western Development Region, Shenzhen Qianhai District, and Zhuhai Hengqin District, which are subject to low income tax rates), and business type (e.g., sales of agricultural products are exempt from VAT; energy-saving and water-saving projects are fully exempt for the first three years after the first operating income is generated, with a 50% reduction for the next three years). Manual selection of applicable rule sets is time-consuming and has limited accuracy.

[0085] On this basis, further, the available rule set obtained in the aforementioned step 201 can be obtained in the following way: input the corporate information of the enterprise to be planned into a pre-built portrait decision tree to determine the corporate portrait; input the corporate portrait into a pre-built rule decision tree to determine the available rule set, which includes multiple available tax calculation rules corresponding to the corporate portrait, and each tax calculation rule includes at least one tax calculation dimension.

[0086] See also Figure 4 Specifically, considering that when manually selecting available rule sets in the existing technology, it is also necessary to judge the available tax calculation rules of the enterprise to be planned based on the enterprise qualifications (i.e., enterprise portrait). Therefore, a portrait decision tree and a rule decision tree are pre-constructed, and the enterprise information of the enterprise to be planned is input into the pre-constructed portrait decision tree to determine the enterprise portrait; and then the enterprise portrait is input into the pre-constructed rule decision tree to obtain the available rule set of the enterprise to be planned. Among them, the enterprise information is the basic information of the enterprise, including but not limited to: the industry to which it belongs, the registration type, the registered address, the asset scale, the revenue scale, the shareholder information, the number of employees and / or other relevant information that can be used to judge the qualifications of the enterprise.

[0087] In practical applications, a portrait decision tree can be constructed based on multiple types of enterprise policies in the enterprise policy database, and / or a rule decision tree can be constructed based on multiple tax calculation rules in the tax policy database.

[0088] It should be noted that enterprise policies include but are not limited to: high-tech enterprise policies, software integrated circuit enterprise policies, technologically advanced service enterprise policies, specific regional enterprise policies, encouraged industry enterprise policies, small and micro-profit enterprise policies, etc.; the generated enterprise portraits may include but are not limited to: high-tech enterprises, small and micro-profit enterprises, integrated circuit manufacturing enterprises, encouraged industry enterprises located in the western region, non-profit organization enterprises, etc.; tax calculation rules include but are not limited to: general value-added tax taxpayer calculation rules, small-scale value-added tax taxpayer calculation rules, value-added tax refund calculation rules, income tax small and micro-profit enterprise calculation rules, income tax high-tech enterprise calculation rules, income tax integrated circuit enterprise calculation rules, income tax specific area calculation rules, small and micro-profit enterprise six taxes and two fees calculation rules, non-profit enterprise calculation rules, etc.

[0089] For example, if Figure 4 The output of the portrait decision tree is: high-tech enterprises, integrated circuit enterprises, small and micro-profit enterprises. Figure 4 The rule decision tree shown may output the following available rule sets: VAT general taxpayer calculation rules, VAT immediate tax refund calculation rules, income tax high-tech enterprise calculation rules, income tax integrated circuit enterprise calculation rules, income tax small and micro-profit enterprise calculation rules, and small and micro-profit enterprise six-tax and two-fee calculation rules. Evaluating these available tax calculation rules, the resulting recommended node set includes the following target available tax calculation rules: VAT general taxpayer calculation rules, VAT immediate tax refund calculation rules, income tax integrated circuit enterprise calculation rules, and small and micro-profit enterprise six-tax and two-fee calculation rules.

[0090] In this embodiment, users only need to input the company information of the company to be planned, and the pre-built portrait decision tree and rule decision tree can determine the applicable rule set corresponding to the company to be planned. This is simple to use, greatly reducing the user's workload and improving tax planning efficiency. In addition, compared to manual screening, the portrait decision tree can accurately judge the company's qualifications (i.e., the company profile) based on the company information, and the rule decision tree can more quickly and accurately screen the available tax calculation rules based on the company profile.

[0091] Furthermore, the user can input at least one rule selection constraint based on the actual situation of the enterprise to be planned. After obtaining the enterprise portrait, the at least one rule selection constraint and the enterprise portrait are input into the rule decision tree to obtain an available rule set. In this way, when determining the available rule set, the rule decision tree will combine at least one rule selection constraint to determine the available rule set. For example, the enterprise to be planned selected the tax calculation rule last month, and tax calculation rule A requires that the enterprise use it for at least half a year. Therefore, the user can input a rule selection constraint to restrict the available rule set output by the rule decision tree to include tax calculation rule A, and at the same time restrict the available rule set output by the rule decision tree to not include any tax calculation rules that cannot be used simultaneously with tax calculation rule A. Among them, the rule selection constraint can be in or outside the enterprise information, and is not limited here.

[0092] It should be noted that when the user is the one providing the enterprise information of the company to be planned, the enterprise information and other related data involved in the embodiments of this application are all obtained after authorization by the user. Furthermore, when the embodiments of this application are applied to specific products or technologies, the data involved must obtain the user's permission or consent, and the collection, use, and processing of the relevant data must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0093] In actual applications, based on the aforementioned embodiments, the available rule set of the embodiments of the present application also includes a mutually exclusive relationship between any two available tax calculation rules. When the aforementioned step 206 determines the recommended node set that meets the preset recommendation conditions based on the path optimization algorithm, the node set in which there is a mutually exclusive relationship between the two available tax calculation rules corresponding to any two target nodes will not be considered, so as to ensure that the recommended tax plan finally determined is a feasible tax plan without a mutually exclusive relationship.

[0094] Based on the above embodiment, in a specific scenario, if the entropy weight method is used to calculate the objective weight of each tax calculation rule in each tax calculation dimension, the relevant steps in the above embodiment can be specifically implemented in the following manner:

[0095] There are m available tax calculation rules: X1, X2, ... X m , each available tax calculation rule corresponds to n tax calculation dimension sets X i ={x1,x2,……x n},based on Normalization is performed, where i represents the i-th tax calculation dimension, j represents the j-th available tax calculation rule, and Y ij is the normalized dimension value of the jth available tax calculation rule in the i-th tax calculation dimension, X ij is the dimension value of the jth available tax calculation rule in the i-th tax calculation dimension, min(X i) is the dimension value with the lowest value under the i-th tax dimension, max(X i ) is the dimension value with the highest value under the i-th tax calculation dimension.

[0096] Calculate the variation size of each available tax calculation rule in each tax calculation dimension, that is, the weight of each available tax calculation rule in each tax calculation dimension: Where i = 1, ..., n; j = 1, ..., m; 0 ≤ p ij ≤1; p ij is the variation size of the j-th available tax calculation rule in the i-th tax calculation dimension.

[0097] Calculate the information entropy of each available tax calculation rule: Among them E j ≥0, and if p ij =0, then define E j =0,E j is the information entropy of the jth available tax calculation rule.

[0098] Calculate the objective weight for each available tax calculation rule: Among them, D j is the degree of difference D of the jth tax calculation rule j =1-E j , w j is the objective weight of the jth available tax calculation rule.

[0099] In addition, the deviation between the historical recommended tax plan and the historical actual tax plan can be calculated as follows:

[0100] A hg =F hg -X hg

[0101] Among them, A hg is the deviation value of the g-th available tax calculation rule under the h-th tax calculation dimension, F ij X is the dimension value of the gth available tax calculation rule under the hth tax calculation dimension in the historical actual tax plan, ij This is the dimension value for the gth available tax calculation rule in the historical recommended tax plan, under the hth tax calculation dimension. Note that if a dimension value does not exist, it can be defined as 0. The available tax calculation rule here can be any available tax calculation rule in the historical recommended tax plan or the historical actual tax plan.

[0102] Then, A hg As X ij And refer to the above steps of this embodiment to calculate and obtain the subjective weight w g ′ , that is, wj ′ .

[0103] Finally, if the jth available tax calculation rule has a subjective weight under a certain tax calculation dimension, then the target weight W of the available tax calculation rule under a certain tax calculation dimension is j =w j +w j ′ , that is, the target weight of the available tax calculation rule in a certain tax calculation dimension is the sum of its corresponding subjective weight and the corresponding objective weight; if the j-th available tax calculation rule does not have a subjective weight in a certain tax calculation dimension, then the target weight of the available tax calculation rule in a certain tax calculation dimension is W j =w j , that is, the target weight of the available tax calculation rule under a certain tax calculation dimension is its objective weight.

[0104] See also Figure 5 The present invention provides an intelligent tax planning device, comprising:

[0105] An acquisition unit 501 is used to acquire an available rule set corresponding to the enterprise to be planned, where the available rule set corresponds to multiple available tax calculation rules;

[0106] A weight calculation unit 502 is configured to determine a dimension value of each available tax calculation rule in each tax calculation dimension in the available rule set, and calculate an objective weight of each tax calculation rule based on the dimension value of each available tax calculation rule in each tax calculation dimension and an objective weight calculation method;

[0107] Determining unit 503 is configured to determine each tax calculation dimension of each available tax calculation rule as a target node, and determine the product of a target weight corresponding to each target node and a dimension value corresponding to each target node, which is the node value of each target node; the target weight corresponding to each target node is the target weight of the available tax calculation rule corresponding to each target node; the target weight of the available tax calculation rule includes the objective weight of the available tax calculation rule;

[0108] The determining unit 503 is further configured to use each target node as a path node and the node value of each target node as the path length, and to determine a group of target nodes whose total path length meets a preset recommendation condition as a recommended node set through a path optimization algorithm;

[0109] The determining unit 503 is further configured to determine a recommended tax plan based on the target available tax calculation rules and target tax calculation dimensions corresponding to each target node in the recommended node set.

[0110] In a specific implementation, the target weights of the available tax calculation rules further include subjective weights of the available tax calculation rules, and the apparatus further includes:

[0111] The acquisition unit 501 is further configured to acquire any historically recommended tax plan and the corresponding historically actual tax plan;

[0112] The weight calculation unit 502 is further used to calculate the deviation value of each applicable tax calculation rule included in the historical recommended tax plan and the historical actual tax plan under each tax calculation dimension;

[0113] The weight calculation unit 502 is further configured to calculate a subjective weight of each available tax calculation rule based on the deviation value of each available tax calculation rule in each tax calculation dimension and the objective weight calculation method;

[0114] The determination unit 503 is further configured to determine a target weight corresponding to each available tax calculation rule, which is the sum of the objective weight and the subjective weight corresponding to each available tax calculation rule.

[0115] In a specific implementation, the determining unit 503 is specifically configured to determine, by using a path optimization algorithm, a group of target nodes with the longest sum of corresponding path lengths, as the recommended node set.

[0116] In a specific implementation, the determination unit 503 is further configured to input the enterprise information of the enterprise to be planned into a pre-built portrait decision tree to determine the enterprise portrait;

[0117] The determination unit 503 is also used to input the enterprise portrait into a pre-built rule decision tree to determine the available rule set, which includes multiple available tax calculation rules corresponding to the enterprise portrait, and each tax calculation rule includes at least one tax calculation dimension.

[0118] In a specific implementation, the apparatus further includes: a construction unit;

[0119] A construction unit, used to construct a profile decision tree based on multiple types of enterprise policies in the enterprise policy database;

[0120] The construction unit is also used to construct a rule decision tree based on multiple tax calculation rules in the tax policy database.

[0121] In a specific implementation, the determining unit 503 is specifically configured to obtain at least one rule selection constraint input by a user;

[0122] At least one rule selection constraint and an enterprise profile are input into the rule decision tree to obtain an available rule set.

[0123] In a specific implementation, the available rule set also includes a mutually exclusive relationship between any two available tax calculation rules; and there is no mutually exclusive relationship between the two available tax calculation rules corresponding to any two target nodes in the recommended node set.

[0124] Figure 6Schematic diagram of the structure of a tax intelligent planning device provided in an embodiment of the present application. The tax intelligent planning device 600 may include one or more central processing units (CPU) 601 and a memory 605. The memory 605 stores one or more applications or data.

[0125] Memory 605 may be volatile or persistent storage. The program stored in memory 605 may include one or more modules, each of which may include a series of instruction operations within the intelligent tax planning device. Furthermore, the central processing unit 601 may be configured to communicate with memory 605 and execute the series of instruction operations within memory 605 on the intelligent tax planning device 600.

[0126] The tax intelligent planning device 600 may also include one or more power supplies 602, one or more wired or wireless network interfaces 603, one or more input and output interfaces 604, and / or one or more operating systems, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0127] The CPU 601 can execute the aforementioned Figures 1 to 5 The operations performed by the intelligent tax planning device in the illustrated embodiment will not be described in detail here.

[0128] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0129] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0130] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0131] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0132] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk, and other media that can store program code.

[0133] An embodiment of the present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the above-mentioned intelligent tax planning method.

Claims

1. A tax intelligent planning method, characterized in that: include: Obtaining an available rule set corresponding to the enterprise to be planned, wherein the available rule set corresponds to multiple available tax calculation rules; Determining a dimension value of each available tax calculation rule in the available rule set in each tax calculation dimension, and calculating an objective weight of each available tax calculation rule in each tax calculation dimension based on the dimension value of each available tax calculation rule in each tax calculation dimension and an objective weight calculation method; Determine each tax calculation dimension of each available tax calculation rule as a target node, and determine the product of a target weight corresponding to each target node and a dimension value corresponding to each target node as the node value of each target node; the target weight corresponding to each target node includes the objective weight of the available tax calculation rule corresponding to each target node under the corresponding tax calculation dimension; Taking each target node as a path node and the node value of each target node as the path length, a group of target nodes whose total path length meets a preset recommendation condition is determined as a recommended node set by a path optimization algorithm; A recommended tax scheme is determined based on the target available tax calculation rules and target tax calculation dimensions corresponding to each target node in the recommended node set; wherein, the total path length of each group of target nodes is used to reflect the comprehensive tax burden of the tax scheme corresponding to each group of target nodes, and the total path length is inversely proportional to the corresponding comprehensive tax burden.

2. The method according to claim 1, characterized in that The target weight corresponding to each target node further includes a subjective weight of the available tax calculation rules corresponding to each target node under the corresponding tax calculation dimension, and the method further includes: Obtain any historical recommended tax plan and the corresponding historical actual tax plan; Calculate the deviation value of each applicable tax calculation rule included in the historical recommended tax plan and the historical actual tax plan under each tax calculation dimension; Based on the deviation value of each available tax calculation rule in each tax calculation dimension and the objective weight calculation method, the subjective weight of each available tax calculation rule in each tax calculation dimension is calculated.

3. The method according to claim 1, characterized in that The group of target nodes whose total path lengths meet the preset recommendation conditions determined by the path optimization algorithm is the recommended node set, including: A set of target nodes with the longest sum of corresponding path lengths is determined by a path optimization algorithm, and is referred to as the recommended node set.

4. The method according to claim 1, wherein The method further comprises: Inputting the enterprise information of the enterprise to be planned into a pre-built portrait decision tree to determine the enterprise portrait; The enterprise portrait is input into a pre-built rule decision tree to determine the available rule set, which includes multiple available tax calculation rules corresponding to the enterprise portrait, and each of the available tax calculation rules includes at least one tax calculation dimension.

5. The method according to claim 4, characterized in that The method further comprises: Constructing the portrait decision tree based on multiple types of enterprise policies in the enterprise policy database; The rule decision tree is constructed based on a plurality of available tax calculation rules in a tax policy database.

6. The method according to claim 4, characterized in that Inputting the enterprise profile into a pre-built rule decision tree to determine the available rule set includes: Obtain at least one rule selection constraint input by a user; The at least one rule selection constraint and the enterprise profile are input into the rule decision tree to obtain the available rule set.

7. The method according to claim 1, characterized in that The available rule set also includes a mutually exclusive relationship between any two available tax calculation rules; and there is no mutually exclusive relationship between the two available tax calculation rules corresponding to any two target nodes in the recommended node set.

8. A tax intelligent planning device, characterized in that: include: An acquisition unit, configured to acquire an available rule set corresponding to the enterprise to be planned, wherein the available rule set corresponds to a plurality of available tax calculation rules; a weight calculation unit, configured to determine a dimension value of each available tax calculation rule in the available rule set in each tax calculation dimension, and calculate an objective weight of each available tax calculation rule in each tax calculation dimension based on the dimension value of each available tax calculation rule in each tax calculation dimension and an objective weight calculation method; a determination unit, configured to determine each tax calculation dimension of each available tax calculation rule as a target node, and determine the product of a target weight corresponding to each target node and a dimension value corresponding to each target node as the node value of each target node; the target weight corresponding to each target node includes an objective weight of the available tax calculation rule corresponding to each target node under the corresponding tax calculation dimension; The determining unit is further configured to use each target node as a path node and the node value of each target node as the path length, and determine, through a path optimization algorithm, a group of target nodes whose total path length meets a preset recommendation condition as a recommended node set; The determination unit is further used to determine a recommended tax plan based on the target available tax calculation rules and target tax calculation dimensions corresponding to each target node in the recommended node set; wherein the total path length of each group of target nodes is used to reflect the comprehensive tax burden of the tax plan corresponding to each group of target nodes, and the total path length is inversely proportional to the corresponding comprehensive tax burden.

9. A tax intelligent planning device, characterized in that: include: CPU, memory and input / output interfaces; The memory is a transient storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the method according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that The computer storage medium stores instructions, which, when executed on a computer, enable the computer to perform the method according to any one of claims 1 to 7.

11. A computer program product comprising instructions, which, when run on a computer, causes the computer to perform the method according to any one of claims 1 to 7.

Images