Multi-dimensional Tax Data Analysis and Prediction Method

Abstract

The present invention provides a multi-dimensional tax data analysis and prediction method, which relates to data processing technology and includes: if a first specific node meets the combination condition, extracting the tax identity information of the corresponding asset tax chain and taking it as the first specific chain; screening the second specific chain corresponding to the tax identity information in the tax database based on the tax identity information, and extracting the second specific node in the second specific chain; if the first specific node and the second specific node correspond to each other, combining them into a combined chain and saving it to the tax database; if it is determined that there is a third specific node in any combined chain, decomposing the combined chain based on the historical first specific node, the historical second specific node and the current third specific node to obtain at least one verification segment, and performing time dimension and value dimension analysis on the verification segment to determine the analysis target end. The present invention can realize the efficient extraction, combination and analysis of asset data, and predict potential tax risks.

Description

Technical Field

[0001] The present invention relates to data processing technologies, and in particular to a multi-dimensional tax data analysis and prediction method. Background Art

[0002] With the complication of the tax management of the entire life cycle of enterprise fixed assets, the tax data brought about by cross-enterprise transfer and multi-stage disposal presents the characteristics of strong temporal correlation and high heterogeneity. For typical fixed assets such as industrial equipment, in the links of purchase, transfer, depreciation, scrapping, etc., discrete tax chain data uploaded by multiple entity tax declaration terminals form a scattered asset transaction track. Tax authorities need to conduct cross-period and cross-entity correlation analysis on a large number of chains to identify risks such as abnormal depreciation rates and inconsistent taxable years. However, problems such as the fragmentation of tax identities and the temporal breakage of transaction nodes generated during multiple transfers of fixed assets make it difficult to trace the entire chain and difficult to achieve dynamic risk warning.

[0003] Existing tax data analysis mostly adopts a combination of static database matching and manual sampling verification. It conducts compliance verification on the asset depreciation data of a single enterprise through preset rules, or conducts retrospective calculation on all transaction nodes within a specific time period during the inspection stage. Such methods have significant defects. They need to rely on manual experience to screen key nodes, cannot automatically identify cross-enterprise data, and lack the multi-dimensional analysis ability for the continuity of depreciation strategies and the consistency of multi-entity transaction logics throughout the life cycle of assets.

[0004] Therefore, how to achieve efficient extraction, combination, and analysis of asset data and predict potential tax risks has become an urgent problem to be solved. Summary of the Invention

[0005] An embodiment of the present invention provides a multi-dimensional tax data analysis and prediction method, which can achieve efficient extraction, combination, and analysis of asset data and predict potential tax risks.

[0006] In a first aspect of an embodiment of the present invention, a multi-dimensional tax data analysis and prediction method is provided, including:

[0007] The server extracts all asset tax chains received within a preset time period, obtains each first specific node in the asset tax chain. If the first specific node meets the combination condition, the tax identity information of the corresponding asset tax chain is extracted and used as the first specific chain;

[0008] Based on the tax identity information, the second specific chain corresponding to the tax identity information in the tax database is screened, and the second specific node in the second specific chain is extracted;

[0009] If the first specific node and the second specific node correspond to each other, they are combined into a combined chain and saved to the tax database;

[0010] If it is determined that there is a third specific node in any of the combined chains, at least one verification segment is obtained by decomposing the combined chain based on the historical first specific node, the historical second specific node, and the current third specific node, and the analysis target end is determined by analyzing the verification segment in terms of time dimension and value dimension.

[0011] Optionally, in a possible implementation manner of the first aspect, the server extracts all asset tax chains received within a preset time period, obtains each first specific node in the asset tax chains. If the first specific node meets the combination condition, the tax identity information of the corresponding asset tax chain is extracted and used as the first specific chain, including:

[0012] The server extracts all asset tax chains received within a preset time period. The asset tax chains are sent by the tax filing end of the enterprise, and each node in the asset tax chains has preset time information;

[0013] The first node in the asset tax chain is extracted as the first specific node. If there is no upload mark for the first specific node, the asset tax chain to which the first specific node belongs is used as the first specific chain that meets the combination condition;

[0014] The attribute information of the first specific node is extracted, and the attribute information includes at least one of the first time information and the first transaction information.

[0015] Optionally, in a possible implementation manner of the first aspect, before the step in which the server extracts all asset tax chains received within a preset time period, obtains each first specific node in the asset tax chains, and if the first specific node meets the combination condition, extracts the tax identity information of the corresponding asset tax chain and uses it as the first specific chain, it further includes:

[0016] After receiving the asset tax chain sent by the tax filing end, the server extracts the tax identity information corresponding to the asset tax chain. Each asset tax chain has preset tax identity information;

[0017] All the pre-stored tax identity information in the tax database is traversed. If it is determined that there is corresponding tax identity information, no upload mark is added to the asset tax chain;

[0018] If it is determined that there is no corresponding tax identity information, an upload mark is added to the asset tax chain.

[0019] Optionally, in a possible implementation manner of the first aspect, the method of screening the second specific chain corresponding to the tax identity information in the tax database based on the tax identity information and extracting the second specific node in the second specific chain includes:

[0020] Determine the last node in the second specific chain as the second specific node, and extract the attribute information of the second specific node, where the attribute information includes at least one of second time information and second transaction information.

[0021] Optionally, in a possible implementation manner of the first aspect, the step of combining the first specific node and the second specific node into a combined chain and saving it to the tax database if they correspond includes:

[0022] Obtain the first transaction feature and the second transaction feature in the first transaction information and the second transaction information, where the first transaction feature and the second transaction feature include at least transaction time, transaction amount, and transaction subject;

[0023] If the first transaction feature and the second transaction feature correspond, determine that the first specific node and the second specific node correspond;

[0024] Connect the second specific node in front and the first specific node behind with a first node connection line in a preset form to obtain a combined chain, where the first node connection line is different from the node connection lines in the first specific chain and the second specific chain;

[0025] Store the combined chain as a new second specific chain in the tax database, and delete the historical second specific chain from the tax database.

[0026] Optionally, in a possible implementation manner of the first aspect, if it is determined that there is a third specific node in any combined chain, at least one verification segment is obtained by decomposing the combined chain based on the historical first specific node, second specific node, and the current third specific node, and the analysis target end is determined by analyzing the verification segment in terms of time dimension and value dimension, including:

[0027] If it is determined that there is a third specific node in any combined chain, the third specific node has scrapping information and / or preset minimum residual value information;

[0028] Perform a splitting process on the combined chain with the first node connection line as the dividing line to obtain multiple split sub-segments;

[0029] Extract the time information of the historical first specific node, second specific node, and the current third specific node, and add processing to the head and tail of each split sub-segment to obtain a verification segment;

[0030] Determine the taxable comprehensive information of the corresponding verification segment based on the header attribute information and the tail attribute information of the verification segment, and perform time dimension and value dimension analysis on each verification segment based on the taxable comprehensive information and tax identity information to determine the analysis target end.

[0031] Optionally, in a possible implementation manner of the first aspect, the determining the taxable comprehensive information of the corresponding verification segment based on the header attribute information and the tail attribute information of the verification segment, and performing time dimension and value dimension analysis on each verification segment based on the taxable comprehensive information and tax identity information to determine the analysis target end includes:

[0032] Determine the preset depreciation life and / or preset depreciation rate of the corresponding entity based on the tax identity information;

[0033] Calculate the time difference between the time in the header attribute information and the time in the tail attribute information of the verification segment to obtain the taxable time period, and calculate the value in the header attribute information and the value in the tail attribute information of the verification segment to obtain the depreciation value;

[0034] Obtain the analysis depreciation rate based on the depreciation value and the taxable time period, calculate the actual life of the initial node and the end node of the combined chain, and perform analysis processing based on the analysis depreciation rate, the actual life, the preset depreciation life and / or the preset depreciation rate to determine the analysis target end.

[0035] Optionally, in a possible implementation manner of the first aspect, the obtaining the analysis depreciation rate based on the depreciation value and the taxable time period, calculating the actual life of the initial node and the end node of the combined chain, and performing analysis processing based on the analysis depreciation rate, the actual life, the preset depreciation life and / or the preset depreciation rate to determine the analysis target end includes:

[0036] If the analysis depreciation rate is greater than the preset depreciation rate, add a first mark to the tax return end that uploads the corresponding verification segment and use it as the first target end;

[0037] If the actual life is less than the preset depreciation life, add a second mark to the tax return end that uploads the corresponding verification segment and use it as the second target end;

[0038] Count the number of times each tax return end serves as the first target end and the second target end, obtain the production data of the tax return end for comprehensive calculation to obtain the analysis coefficient, and if the analysis coefficient is greater than the preset coefficient, use the tax return end as the analysis target end.

[0039] Optionally, in a possible implementation manner of the first aspect, the counting the number of times each tax return end serves as the first target end and the second target end, obtaining the production data of the tax return end for comprehensive calculation to obtain the analysis coefficient, and if the analysis coefficient is greater than the preset coefficient, using the tax return end as the analysis target end includes:

[0040] Count the number of times each tax filing terminal serves as the first target terminal and the second target terminal to obtain the first target count and the second target count. After performing weighted summation processing on the first target count and the second target count respectively, an abnormal sub - coefficient is obtained;

[0041] Statistical production data of the tax filing terminal is input into a pre - configured production evaluation model to obtain a production model coefficient. The production evaluation model includes multiple production units, and each production unit has preset production information and a sub - model;

[0042] Based on an analysis formula, comprehensive calculation is performed on the abnormal sub - coefficient and the production model coefficient to obtain an analysis coefficient.

[0043] Optionally, in a possible implementation manner of the first aspect, the step of inputting the statistical production data of the tax filing terminal into a pre - configured production evaluation model to obtain a production model coefficient includes:

[0044] Decompose the production data into production scale sub - values of multiple production information and then input them into the sub - model to obtain sub - production evaluation values. Each production information has a corresponding production scale sub - value in the numerical range;

[0045] After normalizing all the sub - production evaluation values according to the normalization values corresponding to different production information, they are aggregated to obtain the production model coefficient.

[0046] Optionally, in a possible implementation manner of the first aspect, the analysis coefficient is calculated by the following formula:

[0047] ;

[0048] Where, is the analysis coefficient, is the first target count, is the first weight, is the second target count, is the second weight, is the first constant value, is the preset constant value, is the production scale sub - value of the th production unit in the production evaluation model, is the normalization value of the th production unit in the production evaluation model, is the reference value of the th production unit in the production evaluation model, is the upper limit value of the production unit.

[0049] In a second aspect of the present invention, there is provided a computer device, comprising: a memory, a processor, and a computer program, wherein the computer program is stored in the memory, and the processor runs the computer program to execute the method according to the first aspect of the present invention and various possible methods related to the first aspect.

[0050] In a third aspect of the present invention, there is provided a storage medium, in which a computer program is stored, and when the computer program is executed by a processor, it is used to implement the method according to the first aspect of the present invention and various possible methods related to the first aspect.

[0051] The technical solution of the present invention can achieve the following technical effects:

[0052] Efficient chain extraction based on a preset time period and combined conditions. The present invention constructs a multi-level chain association system through a tax identity identification and verification mechanism, a first node marking strategy, and a differential node connection line design. Through the preset time period and combined conditions, efficient extraction of the tax chain of newly added assets is achieved. Specifically, the system extracts all asset tax chains according to the preset time period, and determines whether the combined conditions are met by scanning the first specific node (i.e., the starting node of the chain) of each chain. If the first specific node has no historical mark, it is marked as the first specific chain, and the corresponding tax identity information is extracted. This technical effect solves the problem of low data processing efficiency in the prior art, and significantly reduces the data processing volume by quickly locating newly added business chains.

[0053] Chain combination based on tax identity information and historical data. The present invention realizes the intelligent combination of asset tax chains through tax identity information and historical data. Specifically, the system filters the second specific chains in the tax database based on tax identity information, and extracts its last node (i.e., the second specific node). If the transaction characteristics (such as transaction time, transaction amount, transaction entity) of the first specific node and the second specific node match, the two are combined into a new combined chain and saved to the tax database. This technical effect solves the problem of single analysis dimension in the prior art, and significantly improves the accuracy and integrity of chain combination through the matching of multi-dimensional transaction characteristics.

[0054] Tax risk prediction based on time dimension and value dimension. Through the analysis of time dimension and value dimension, the present invention realizes the accurate prediction of potential tax risks. Specifically, when a third specific node (such as an equipment scrapping node or a residual value assessment node) is detected in the combined chain, the system automatically triggers the intelligent decomposition of the chain and generates a verification segment covering the three stages of purchase, holding, and disposal. Then, based on the head attribute information and tail attribute information of the verification segment, the taxable time period and depreciation value are calculated, and analysis is carried out in combination with the preset depreciation life and depreciation rate to determine the analysis target end. This technical effect solves the problem of the lack of dynamic prediction ability in the prior art, and significantly improves the accuracy and practicality of tax risk prediction through multi-dimensional analysis and dynamic weight adjustment mechanism. Brief Description of the Drawings

[0055] Figure 1 is a schematic flowchart of a multi-dimensional tax data analysis and prediction method provided by an embodiment of the present invention;

[0056] Figure 2 is a schematic hardware structure diagram of a computer device provided by an embodiment of the present invention. Detailed Embodiments

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0058] See Figure 1 , which is a schematic flowchart of a multi-dimensional tax data analysis and prediction method provided by an embodiment of the present invention. The method includes:

[0059] S1. The server extracts all asset tax chains received within a preset time period, obtains each first specific node in the asset tax chain. If the first specific node meets the combination condition, the tax identity information of the corresponding asset tax chain is extracted and used as the first specific chain.

[0060] In this step, the asset tax chain refers to the tax data stream uploaded by the enterprise through the tax reporting terminal, which consists of multiple transaction nodes with timestamps. It should be noted that an asset tax chain corresponds to a fixed asset of the enterprise, such as an industrial equipment. Since a fixed asset will be involved in a series of subsequent disposals, such as purchase, transfer, scrapping, etc., and the determination of tax data is required during the disposal, therefore, this solution sets up an asset tax chain for recording, which is used for subsequent tax verification and determination of risk data for reminder.

[0061] The server first grabs all the chain data within a preset time period (such as the second quarter of 2024), and determines whether it meets the combination conditions by scanning the first specific node (i.e., the starting node of the chain) of each chain. The system identifies that there is no historical mark on its first node, indicating that it is newly constructed, then marks the whole chain as the first specific chain, and extracts the tax identity identifier of the fixed asset. Among them, the tax identity identifier can be pre-configured, and each device corresponds to one. This step realizes the rapid positioning of the new business chain, reducing the data processing volume compared with the traditional full-scale retrieval method.

[0062] In some embodiments, the server extracts all the asset tax chains received within a preset time period, obtains each first specific node in the asset tax chain. If the first specific node meets the combination conditions, then extracts the tax identity information of the corresponding asset tax chain and takes it as the first specific chain, including:

[0063] S11, the server extracts all the asset tax chains received within a preset time period. The asset tax chain is sent by the enterprise's tax reporting terminal, and each node in the asset tax chain has preset time information.

[0064] Among them, the preset time period in this step refers to a configurable tax analysis period (such as a natural month, quarter, or custom time period). The server receives the encrypted data stream sent by the enterprise's tax reporting terminal through the API interface. Each node of each asset tax chain carries machine-readable time information. This design ensures data timeliness and traceability, and establishes a standardized data structure for subsequent time series analysis.

[0065] S12, extracts the first node in the asset tax chain as the first specific node. If there is no upload mark on the first specific node, then takes the asset tax chain to which the first specific node belongs as the first specific chain that meets the combination conditions.

[0066] Among them, the upload mark here can refer to the identifier (0 / 1 value) implanted by the server in the node metadata. The system preferentially detects whether the first node of the chain contains an upload mark with a value of 0 (such as the tax chain declared by an enterprise for the first time). If the condition is met, the chain capture mechanism is triggered. For example, when an enterprise first uploads the purchase tax data of equipment A in June 2024, the first node of its chain is marked as 0 due to no historical record. At this time, this solution will use it as the first specific chain for subsequent processing.

[0067] S13. Extract the attribute information of the first specific node, where the attribute information includes at least one of the first time information and the first transaction information.

[0068] It can be understood that this solution will capture the relevant information of the first specific node for subsequent processing. Among them, the first time information is the occurrence time of this node, such as the time of the transaction, and the first transaction information refers to what transaction is done for this fixed asset, such as purchase.

[0069] In some embodiments, before the step of the server extracting all asset tax chains received within a preset time period, obtaining each first specific node in the asset tax chain, and if the first specific node meets the combination condition, extracting the tax identity information of the corresponding asset tax chain and using it as the first specific chain, the following steps are further included:

[0070] After the server receives the asset tax chain sent by the tax reporting end, it extracts the tax identity information corresponding to the asset tax chain, and each asset tax chain has a preset tax identity information.

[0071] The tax identity information in this step can be the unique identity identifier corresponding to the fixed asset issued by the tax authority. When the server receives the asset tax chain transmitted by the tax reporting end (such as the electronic tax bureau client), it will extract the tax identity information corresponding to the chain, and the tax identity information of different assets is different.

[0072] Traverse all the tax identity information prestored in the tax database. If it is judged that there is corresponding tax identity information, no upload mark is added to the asset tax chain.

[0073] Subsequently, traverse the tax database for comparison: If there is the same identity record, skip the marking operation. It can be understood that if it exists in the database, it means that the chain has been uploaded before and there is no need to upload it.

[0074] If it is judged that there is no corresponding tax identity information, an upload mark is added to the asset tax chain.

[0075] If there is no match, an upload mark with a value of 0 is inserted into the chain metadata. It can be understood that if it does not exist in the database, it means that it has not been uploaded before and needs to be uploaded.

[0076] S2. Based on the tax identity information, filter the second specific chain corresponding to the tax identity information in the tax database, and extract the second specific node in the second specific chain.

[0077] It can be understood that fixed assets may be involved in transactions such as transfers. Therefore, there may be corresponding chains in the database for a fixed asset. For example, current enterprise B uploads the tax chain of device 1, and a certain enterprise A has also uploaded the tax chain of device 1 before, but enterprise A sold device 1 to enterprise B.

[0078] In some embodiments, the filtering the second specific chain corresponding to the tax identity information in the tax database based on the tax identity information and extracting the second specific node in the second specific chain includes:

[0079] S21. Determine the last node in the second specific chain as the second specific node, and extract the attribute information of the second specific node. The attribute information includes at least one of the second time information and the second transaction information.

[0080] Among them, the last node is defined here as the node with the latest timestamp in the second specific chain. The second time information extracted by the system includes the node occurrence time, and the second transaction information covers transaction information such as transaction time, transaction amount, and transaction subject.

[0081] S3. If the first specific node and the second specific node correspond to each other, combine them into a combined chain and save it to the tax database.

[0082] This solution will combine the first specific node and the second specific node for combined judgment. When the combination conditions are met, the first specific chain and the second specific chain will be combined to generate a new summary chain and saved to the tax database.

[0083] In some embodiments, the combining them into a combined chain and saving it to the tax database if the first specific node and the second specific node correspond to each other includes:

[0084] S31. Obtain the first transaction feature and the second transaction feature in the first transaction information and the second transaction information. The first transaction feature and the second transaction feature include at least transaction time, transaction amount, and transaction subject.

[0085] First of all, this solution needs to extract the first transaction information and the second transaction information, parse them, and obtain the corresponding transaction time, transaction amount, and transaction subject.

[0086] S32. If the first transaction feature and the second transaction feature correspond to each other, then determine whether the first specific node and the second specific node correspond to each other.

[0087] Since the first specific node is at the front of the first specific chain and the second specific node is at the end of the second specific chain, if the transaction time, transaction amount, and transaction entity all correspond, it indicates that the two can be connected and combined. For example, when enterprise B uploads the tax chain of device 1, the system detects that its purchase transaction feature exactly matches the tax chain of device 1 transferred by enterprise A, thus triggering the chain combination operation. Among them, entity correspondence can be to determine whether the transferring entities correspond. For example, enterprise B transfers to enterprise A, and enterprise A purchases from enterprise B, which can be considered entity correspondence.

[0088] S33. Connect the second specific node in the front and the first specific node in the back with a first node connection line in a preset form to obtain a combined chain, and the first node connection line is different from the node connection lines within the first specific chain and the second specific chain.

[0089] Since the second specific node corresponds to historical time and the first specific node corresponds to the current time, therefore, when combining, this solution needs to connect the second specific node in the front and the first specific node in the back with a first node connection line in a preset form to obtain a combined chain.

[0090] Among them, the first node connection line in the preset form is different from the node connection lines within the first specific chain and the second specific chain. The part of the first node connection line connected to the first specific chain can be a red line, and the part of the first node connection line connected to the second specific chain can be a green line. Through the above display form, the combination position can be visually displayed.

[0091] S34. Store the combined chain as the new second specific chain in the tax database, and delete the historical second specific chain from the tax database.

[0092] It can be understood that after obtaining the combined chain, the original second specific chain can be deleted from the tax database.

[0093] S4. If it is determined that there is a third specific node in any of the combined chains, then decompose the combined chain based on the historical first specific node, historical second specific node, and current third specific node to obtain at least one verification segment, and perform time - dimension and value - dimension analysis on the verification segment to determine the analysis target end.

[0094] This step realizes the tax verification of the full life cycle of fixed assets. When it is detected that there is a third specific node (such as an equipment scrapping node or a residual value evaluation node) in the combined chain, the intelligent decomposition of the chain is automatically triggered. For example, enterprise B uploads the scrapping tax chain of equipment 1 (with a residual value rate of 5%). The system traces back the historical chain of this equipment (including the transfer by enterprise A, the purchase by enterprise B, and multiple depreciation nodes), and generates a verification segment covering the three stages of purchase, holding, and disposal. Finally, the analysis target end is determined by analyzing the verification segment in terms of time dimension and value dimension.

[0095] In some embodiments, if it is determined that there is a third specific node in any combined chain, then at least one verification segment is obtained by decomposing the combined chain based on the historical first specific node, second specific node, and the current third specific node, and the analysis target end is determined by analyzing the verification segment in terms of time dimension and value dimension, including:

[0096] S41, if it is determined that there is a third specific node in any combined chain, the third specific node has scrapping information and / or preset minimum residual value information.

[0097] Among them, the determination criteria of the third specific node include multiple threshold conditions, including scrapping information and / or preset minimum residual value information. Among them, the scrapping information means that the fixed asset is to be scrapped, for example, the value is 0; the preset minimum residual value information, for example, 5% of the total price. It can be understood that after reaching the above node, it is necessary to trace back the tax chain to see if there are any non-compliant places in the corresponding enterprise tax treatment. Among them, the information of the third specific node can be uploaded during asset transactions.

[0098] S42, the combined chain is split using the first node connection line as the dividing line to obtain multiple split sub-segments.

[0099] First of all, this solution needs to use the first node connection line as the dividing line to split the combined chain to obtain multiple split sub-segments. It can be understood that after the above splitting, 1 split sub-segment can correspond to 1 enterprise entity.

[0100] S43, extract the time information of the historical first specific node, second specific node, and the current third specific node, and add processing to the head and tail of each split sub-segment to obtain a verification segment.

[0101] After obtaining multiple split sub-segments, this solution needs to further process them. It is necessary to add time to the head and tail of each split sub-segment for processing to obtain a verification segment. The first specific node can be the start of time, for example, January 2010, and the second specific node can be the end of time, for example, January 2011. Then the first verification segment can be from January 2010 to January 2011.

[0102] S44. Determine the taxable comprehensive information corresponding to the verification segment based on the header attribute information and the tail attribute information of the verification segment. Analyze each verification segment in terms of time dimension and value dimension based on the taxable comprehensive information and the tax identity information to determine the analysis target end.

[0103] Based on the header attribute information and the tail attribute information of the verification segment, calculate the taxable comprehensive information (such as taxable time period, depreciation value, etc.). Then, in combination with the tax identity information (such as preset depreciation life, preset depreciation rate, etc.), analyze the verification segment in terms of time dimension and value dimension, and finally determine the analysis target end (such as whether tax declaration needs to be adjusted, whether there are tax risks, etc.).

[0104] Among them, the determining the taxable comprehensive information corresponding to the verification segment based on the header attribute information and the tail attribute information of the verification segment, and analyzing each verification segment in terms of time dimension and value dimension based on the taxable comprehensive information and the tax identity information to determine the analysis target end includes:

[0105] S441. Determine the preset depreciation life and / or preset depreciation rate of the corresponding entity based on the tax identity information.

[0106] Among them, the preset depreciation life refers to the pre-set asset depreciation life according to tax policies or accounting standards. The preset depreciation rate refers to the pre-set asset depreciation rate according to tax policies or accounting standards. This solution needs to determine the preset depreciation life and / or preset depreciation rate of the corresponding entity based on the tax identity information. These preset values are the basis for subsequent calculations and are used to judge whether the actual depreciation meets tax requirements. For example, the tax identity information of an asset shows that it belongs to a vehicle, and the applicable preset depreciation life is 10 years and the preset depreciation rate is 10%. These information will be used for subsequent depreciation value calculation and analysis.

[0107] S442. Calculate the time difference between the time in the header attribute information of the verification segment and the time in the tail attribute information to obtain the taxable time period, and calculate the value in the header attribute information of the verification segment and the value in the tail attribute information to obtain the depreciation value.

[0108] Among them, the taxable period refers to the difference between the start time and the end time of the verification segment, representing the time range for which tax needs to be calculated. The depreciation value refers to the difference between the start value and the end value of the verification segment, representing the depreciation amount of the asset during this period. The taxable period is calculated based on the start time and the end time of the verification segment; the depreciation value is calculated based on the start value and the end value of the verification segment. These data are the basis for subsequent analysis. For example, if the start time of a verification segment is January 1, 2020, and the end time is January 1, 2023, the taxable period is 3 years; the start value is 1 million yuan, and the end value is 700,000 yuan, then the depreciation value is 300,000 yuan. By accurately calculating the taxable period and the depreciation value, the accuracy of the analysis results is ensured.

[0109] S443. Obtain an analysis depreciation rate based on the depreciation value and the taxable period, calculate the actual service life of the initial node and the end node of the combined chain, and perform analysis processing based on the analysis depreciation rate, the actual service life, the preset depreciation service life, and / or the preset depreciation rate to determine the analysis target end.

[0110] The analysis depreciation rate refers to the actual depreciation rate calculated based on the depreciation value and the taxable period. The actual service life refers to the time difference between the initial node and the end node of the combined chain. The analysis target end refers to the target result determined through analysis, such as whether tax declaration needs to be adjusted and whether there are tax risks. This solution will calculate the analysis depreciation rate based on the depreciation value and the taxable period; calculate the actual service life based on the initial node and the end node of the combined chain. Then, compare and analyze the analysis depreciation rate, the actual service life with the preset depreciation service life and / or the preset depreciation rate, and finally determine the analysis target end. Through the comparison and analysis, it is determined whether tax declaration needs to be adjusted or whether there are tax risks. Enhance tax compliance and ensure that depreciation calculation and analysis comply with tax policy requirements.

[0111] This solution can improve the efficiency of tax verification. Through automated calculation and analysis, it reduces manual intervention. Ensure that depreciation calculation and analysis comply with tax policy requirements. Provide a decision-making basis. By analyzing the depreciation rate and the actual service life, it is determined whether tax declaration needs to be adjusted or whether there are tax risks. Enhance data analysis capabilities. Through analysis in the time dimension and the value dimension, it provides a comprehensive tax verification result.

[0112] In some embodiments, the obtaining an analysis depreciation rate based on the depreciation value and the taxable period, calculating the actual service life of the initial node and the end node of the combined chain, and performing analysis processing based on the analysis depreciation rate, the actual service life, the preset depreciation service life, and / or the preset depreciation rate to determine the analysis target end includes:

[0113] S4431. If the analysis depreciation rate is greater than the preset depreciation rate, add a first mark to the tax reporting end that uploads the corresponding verification segment and use it as the first target end.

[0114] This solution will first obtain an analysis depreciation rate based on the depreciation value of the asset and the taxable time period. Then, by analyzing this depreciation rate and the actual age of the asset, judgments and further analyses are carried out.

[0115] For example, assume that an enterprise purchases a piece of equipment with an original value of 1 million yuan and a planned service life of 10 years (this is the "preset depreciation life"). During the depreciation process, if the enterprise depreciates at the depreciation rate stipulated by the tax department (for example, 10%), but the actual depreciation rate of the equipment during use is 12% (analysis depreciation rate). At this time, the analysis depreciation rate (12%) is greater than the preset depreciation rate (10%), then the tax return end of this equipment will be marked as the first target end because the depreciation speed is faster. There may be non-compliant places.

[0116] S4432, if the actual age is less than the preset depreciation life, then add a second mark to the tax return end that uploads the corresponding verification segment and use it as the second target end.

[0117] Assume that the equipment has been used for 5 years and the preset depreciation life is 10 years. Then the actual age is less than the preset depreciation life, and the tax return end of the equipment will be marked as the second target end.

[0118] S4433, count the number of times each tax return end is used as the first target end and the second target end, obtain the production data of the tax return end for comprehensive calculation to get the analysis coefficient. If the analysis coefficient is greater than the preset coefficient, then use the tax return end as the analysis target end.

[0119] Finally, conduct statistical analysis on the tax return end of each piece of equipment, and comprehensively calculate the analysis coefficient based on various data. If the analysis coefficient is greater than a certain standard value, then the tax return end of this equipment will be confirmed as the analysis target end, that is, enter the next step of audit or processing. This analysis method can help enterprises accurately judge and mark different tax return ends, making the calculation of depreciation and data analysis more accurate. By analyzing multiple factors such as the depreciation rate, actual age, and preset depreciation life, the depreciation strategy of assets can be better discovered. At the same time, this technology can improve the efficiency of tax audits, avoid tax disputes caused by improper depreciation strategies, and ensure that enterprises conduct reasonable depreciation accounting within the framework of compliance. The implementation of this technology is not only applicable to the asset management and tax return optimization within enterprises, but also can improve the ability to identify and process depreciation data during tax audits, thereby effectively reducing tax risks.

[0120] Among them, the counting of the number of times each tax return end is used as the first target end and the second target end, obtaining the production data of the tax return end for comprehensive calculation to get the analysis coefficient. If the analysis coefficient is greater than the preset coefficient, then use the tax return end as the analysis target end, includes:

[0121] The number of times each tax filing terminal serves as the first target terminal and the second target terminal is counted to obtain the first target count and the second target count. After weighted summation processing of the first target count and the second target count, an abnormal sub - coefficient is obtained.

[0122] This solution will count, for each tax filing terminal, the number of times it serves as the first target terminal and the second target terminal. Calculate the production data of each terminal and synthesize the data to obtain an analysis coefficient. If this coefficient is greater than a certain standard value, the tax filing terminal is regarded as an analysis target terminal.

[0123] Specifically, this solution will count, for each tax filing terminal, the number of times it is marked as the first target terminal and the second target terminal, respectively obtaining the first target count and the second target count. Weights are assigned to the first target count and the second target count respectively (the weights can be set according to the actual situation). The weighted first target count and the second target count are added together to obtain the abnormal sub - coefficient.

[0124] The production data of the tax filing terminal is counted and input into a pre - configured production evaluation model to obtain a production model coefficient. The production evaluation model includes multiple production units, and each production unit has preset production information and sub - models.

[0125] Count the production data of the tax filing terminal (such as personnel, output, income, etc.). Input the production data into a pre - configured production evaluation model. The production evaluation model consists of multiple production units, and each production unit contains preset production information and sub - models. The production model coefficient is calculated through the production evaluation model. Among them, the production information and sub - models can correspond to dimensions such as the personnel dimension, the total assets dimension, etc. It can be understood that the larger the scale of the enterprise, the greater the unreasonable space given by this solution, and vice versa, the smaller the unreasonable space.

[0126] Based on the analysis formula, comprehensive calculation is performed on the abnormal sub - coefficient and the production model coefficient to obtain the analysis coefficient.

[0127] Finally, using the analysis formula, comprehensive calculation is performed on the abnormal sub - coefficient and the production model coefficient to obtain the analysis coefficient. If the analysis coefficient is greater than the preset coefficient, the tax filing terminal is marked as an analysis target terminal and enters the subsequent analysis or processing process. This solution can quickly identify abnormal tax filing terminals through counting times and weighted calculation. Combining with the production evaluation model, it can more comprehensively evaluate whether the production activities of the tax filing terminal are reasonable. Through the preset coefficient and the analysis formula, the analysis criteria can be flexibly adjusted according to the actual situation. The automated calculation and judgment process reduces the influence of human errors and subjective factors. This method can effectively improve the efficiency and accuracy of tax audits, and at the same time provide more scientific tax management support for enterprises.

[0128] Among them, the production data of the statistical tax filing terminal is input into a pre-configured production evaluation model to obtain a production model coefficient, including:

[0129] After decomposing the production data into sub-production scale values of multiple production information and inputting them into the sub-model to obtain sub-production evaluation values, each production information has a corresponding sub-production scale value for its numerical range.

[0130] Decompose the production data of the tax filing terminal into sub-production scale values of multiple production information. For example, the production data can include production volume, cost, revenue, etc., and each indicator corresponds to a sub-production scale value. Input each sub-production scale value into the corresponding sub-model to obtain sub-production evaluation values. Each sub-model may be an independent calculation module for evaluating the rationality or abnormality of this production information. The larger the enterprise scale, the larger the corresponding sub-production scale value. For example, the production information can include the number of people, the number of production equipment, electricity consumption, and so on.

[0131] After normalizing all the sub-production evaluation values according to the normalization values corresponding to different production information, summarize them to obtain the production model coefficient.

[0132] Among them, the purpose of normalization is to convert sub-production evaluation values with different dimensions or ranges into a unified scale for subsequent summarization. Summarize all the normalized sub-production evaluation values to obtain the production model coefficient.

[0133] In the above embodiment, the analysis coefficient is calculated through the following formula

[0134] ;

[0135] Among them, is the analysis coefficient, is the first target number of times, is the first weight, is the second target number of times, is the second weight, is the first constant value, is the preset constant value, is the sub-production scale value of the th production unit in the production evaluation model, is the normalization value of the th production unit in the production evaluation model, is the benchmark value of the th production unit in the production evaluation model, is the upper limit value of the production unit.

[0136] In the above formula, the numerator mainly reflects the degree of abnormality at the tax filing end. The denominator mainly reflects the comprehensive performance of the production data at the tax filing end. The larger the numerator (the higher the degree of abnormality) and the smaller the denominator (the worse the performance of the production data, for example, the smaller the enterprise scale), the larger the analysis coefficient. If it is greater than the preset coefficient, the tax filing end will be marked as the analysis target end.

[0137] By decomposing the production data and inputting it into the sub-model, it is possible to conduct a refined evaluation of each production unit. The normalization process ensures the comparability of data from different production units. The weights, constant values, and reference values in the formula can be dynamically adjusted according to actual needs to adapt to different analysis scenarios. Through the comprehensive calculation of the numerator and denominator, it is possible to comprehensively reflect the degree of abnormality at the tax filing end and the performance of the production data. This method can effectively improve the accuracy and efficiency of tax audits, and at the same time provide more scientific tax management support for enterprises.

[0138] See Figure 2 , which is a schematic diagram of the hardware structure of a computer device provided by an embodiment of the present invention. The computer device 20 includes: a processor 21, a memory 22, and a computer program; where

[0139] The memory 22 is used to store the computer program, and this memory can also be a flash memory. The computer program is, for example, an application program or a functional module that implements the above method.

[0140] The processor 21 is used to execute the computer program stored in the memory to implement each step executed by the device in the above method. Specifically, reference can be made to the relevant descriptions in the foregoing method embodiments.

[0141] Optionally, the memory 22 can be either independent or integrated with the processor 21.

[0142] When the memory 22 is a device independent of the processor 21, the device may further include:

[0143] A bus 23 for connecting the memory 22 and the processor 21.

[0144] The present invention also provides a storage medium in which a computer program is stored, and when the computer program is executed by a processor, it is used to implement the methods provided by the above various embodiments.

[0145] Among them, the storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general or special-purpose computer. For example, the storage medium is coupled to the processor, so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application specific integrated circuit (ASIC). In addition, the ASIC can be located in the user equipment. Of course, the processor and the storage medium can also exist as discrete components in the communication device. The storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0146] The present invention also provides a program product, which includes execution instructions stored in a storage medium. At least one processor of the device can read the execution instructions from the storage medium, and the execution of the execution instructions by at least one processor enables the device to implement the methods provided by the above various embodiments.

[0147] In the above embodiments of the terminal or the server, it should be understood that the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the present invention can be directly embodied as being executed and completed by a hardware processor, or can be executed and completed by a combination of hardware and software modules in the processor.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A multi-dimensional tax data analysis and prediction method, characterized in that Including: The server extracts all asset tax chains received within a preset time period, obtains each first specific node in the asset tax chain. If the first specific node meets the combination condition, the tax identity information of the corresponding asset tax chain is extracted and used as the first specific chain, including: The server extracts all asset tax chains received within a preset time period. The asset tax chains are sent by the tax filing end of the enterprise, and each node in the asset tax chain has preset time information; The first node in the asset tax chain is extracted as the first specific node. If there is no upload mark for the first specific node, the asset tax chain to which the first specific node belongs is used as the first specific chain that meets the combination condition; The attribute information of the first specific node is extracted. The attribute information includes at least one of the first time information and the first transaction information; Based on the tax identity information, the second specific chain corresponding to the tax identity information in the tax database is screened, and the second specific node in the second specific chain is extracted, including: The last node in the second specific chain is determined as the second specific node, and the attribute information of the second specific node is extracted. The attribute information includes at least one of the second time information and the second transaction information; If the first specific node and the second specific node correspond to each other, they are combined into a combined chain and saved to the tax database; If it is determined that there is a third specific node in any of the combined chains, at least one verification segment is obtained by decomposing the combined chain based on the historical first specific node, the historical second specific node, and the current third specific node. The time dimension and value dimension of the verification segment are analyzed to determine the analysis target end, including: If it is determined that there is a third specific node in any of the combined chains, the third specific node has scrapping information and / or preset minimum residual value information; The combined chain is split based on the first node connection line to obtain multiple split sub-segments; The time information of the historical first specific node, the second specific node, and the current third specific node is extracted to add processing to the head and tail of each split sub-segment to obtain a verification segment; Based on the head attribute information and the tail attribute information of the verification segment, the taxable comprehensive information of the corresponding verification segment is determined. Based on the taxable comprehensive information and the tax identity information, the time dimension and value dimension of each verification segment are analyzed to determine the analysis target end.

2. The multi-dimensional tax data analysis and prediction method according to claim 1, characterized in that Before the step in which the server extracts all asset tax chains received within a preset time period, obtains each first specific node in the asset tax chain, and if the first specific node meets the combination condition, extracts the tax identity information of the corresponding asset tax chain and uses it as the first specific chain, it further includes: After the server receives the asset tax chain sent by the tax filing end, it extracts the tax identity information corresponding to the asset tax chain. Each asset tax chain has a preset tax identity information; Traverse all the pre-stored tax identity information in the tax database. If it is determined that there is corresponding tax identity information, do not add an upload mark to the asset tax chain; If it is determined that there is no corresponding tax identity information, add an upload mark to the asset tax chain.

3. The multi-dimensional tax data analysis and prediction method according to claim 1, wherein: When the first specific node and the second specific node correspond to each other and are combined into a combined chain and saved to the tax database, it includes: Obtain the first transaction feature and the second transaction feature in the first transaction information and the second transaction information. The first transaction feature and the second transaction feature at least include transaction time, transaction amount, and transaction entity; If the first transaction feature and the second transaction feature correspond to each other, determine that the first specific node and the second specific node correspond to each other; Connect the second specific node in front and the first specific node at the back with a first node connection line in a preset form to obtain a combined chain. The first node connection line is different from the node connection lines in the first specific chain and the second specific chain; Store the combined chain as the new second specific chain in the tax database, and delete the historical second specific chain from the tax database.

4. The multi-dimensional tax data analysis and prediction method according to claim 1, wherein: Based on the header attribute information and the tail attribute information of the verification segment, determine the taxable comprehensive information of the corresponding verification segment. Based on the taxable comprehensive information and the tax identity information, analyze each verification segment in the time dimension and the value dimension to determine the analysis target end, including: Based on the tax identity information, determine the preset depreciation life and / or preset depreciation rate of the corresponding entity; Calculate the time difference between the time in the header attribute information of the verification segment and the time in the tail attribute information to obtain the taxable time period, and calculate the value in the header attribute information of the verification segment and the value in the tail attribute information to obtain the depreciation value; Based on the depreciation value and the taxable time period, obtain the analysis depreciation rate, calculate the actual life of the initial node and the end node of the combined chain, and perform analysis processing based on the analysis depreciation rate, the actual life, the preset depreciation life and / or the preset depreciation rate to determine the analysis target end.

5. The multi-dimensional tax data analysis and prediction method according to claim 4, wherein: Based on the depreciation value and the taxable time period, obtain the analysis depreciation rate, calculate the actual life of the initial node and the end node of the combined chain, and perform analysis processing based on the analysis depreciation rate, the actual life, the preset depreciation life and / or the preset depreciation rate to determine the analysis target end, including: If the analysis depreciation rate is greater than the preset depreciation rate, add a first mark to the tax return end for uploading the corresponding verification segment and use it as the first target end; If the actual life is less than the preset depreciation life, add a second mark to the tax return end for uploading the corresponding verification segment and use it as the second target end; Count the number of times each tax return end serves as the first target end and the second target end, obtain the production data of the tax return end for comprehensive calculation to obtain the analysis coefficient. If the analysis coefficient is greater than the preset coefficient, use the tax return end as the analysis target end.

6. The multi-dimensional tax data analysis and prediction method according to claim 5, wherein the counting of the number of times each tax filing terminal serves as the first target terminal and the second target terminal, obtaining the production data of the tax filing terminal and comprehensively calculating to obtain an analysis coefficient, and if the analysis coefficient is greater than a preset coefficient, then taking the tax filing terminal as an analysis target terminal, includes: counting the number of times each tax filing terminal serves as the first target terminal and the second target terminal to obtain a first target number and a second target number, and respectively performing weighted summation processing on the first target number and the second target number to obtain an abnormal sub-coefficient; counting the production data of the tax filing terminal input into a pre-configured production evaluation model to obtain a production model coefficient, the production evaluation model includes multiple production units, and each production unit has preset production information and a sub-model; comprehensively calculating the abnormal sub-coefficient and the production model coefficient based on an analysis formula to obtain an analysis coefficient.

7. The multi-dimensional tax data analysis and prediction method according to claim 6, wherein the counting of the production data of the tax filing terminal input into a pre-configured production evaluation model to obtain a production model coefficient, includes: decomposing the production data into production scale sub-values of multiple production information and inputting them into the sub-model to obtain sub-production evaluation values, and each production information has a corresponding production scale sub-value for its numerical range; after normalizing all the sub-production evaluation values according to the normalization values corresponding to different production information, summarizing to obtain a production model coefficient.

8. The multi-dimensional tax data analysis and prediction method according to claim 7, wherein the analysis coefficient is calculated by the following formula ， Among them, is the analysis coefficient, is the first target number of times, is the first weight, is the second target number of times, is the second weight, is the first constant value, is the preset constant value, is the production scale sub-value of the th production unit in the production evaluation model, is the normalized value of the th production unit in the production evaluation model, is the reference value of the th production unit in the production evaluation model, is the upper limit value of the production unit.

9. Computer device, characterized in that, including: a memory, a processor, and a computer program, the computer program is stored in the memory, and the processor runs the computer program to execute the method according to any one of claims 1 to 8.

10. Storage medium, characterized in that, A computer program is stored in the storage medium, and when the computer program is executed by a processor, it is used to implement the method according to any one of claims 1 to 8.

Images