A classification storage method for information technology data

Abstract

The present invention discloses a classification storage method for information technology data, and specifically relates to the technical field of online classification storage. The specific method includes the following steps: representing historical financial data in the form of a knowledge graph, obtaining the correlation between real-time financial data and historical financial data, and obtaining the timeliness of the historical financial data based on the timestamp of the historical financial data, comprehensively analyzing the correlation between the real-time financial data and the historical financial data and the timeliness of the historical financial data to obtain reference data that can be used as reference data for current financial institutions, merging the obtained reference data with the real-time financial data for analysis to obtain the fluctuation of financial amounts, the frequency information of transactions and the speed information of a single transaction, and determining the financial data that needs to be stored separately in the merged financial data, which helps financial institutions to screen historical financial data in other financial institutions, retain financial information that is meaningful to customers, and more accurately locate financial information with potential risks.

Description

Technical Field

[0001] This invention relates to the field of online classification and storage technology, and more specifically, to a method for classifying and storing information technology data. Background Technology

[0002] Currently, financial institutions can obtain customers' financial information from other financial institutions using standardized information exchange methods. However, given the differences in assessment indicators and models used by different financial institutions to evaluate customer financial information, financial institutions usually still need to reassess customer financial information to ensure its accuracy. Therefore, it is not easy to determine whether a customer's historical financial information from other financial institutions can be stored as reference data in the current financial institution's customer financial file. Furthermore, it is not easy for financial institutions to determine which financial information in a customer's financial file has potential risks. Financial institutions need to separately classify and store financial information with potential risks to help professionals analyze it.

[0003] To address the aforementioned shortcomings, a technical solution is provided. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method for classifying and storing information technology data to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for classifying and storing information technology data includes the following steps:

[0007] S1: Transfer the customer's financial information from other financial institutions to the current financial institution, and re-evaluate the customer's financial information at the current financial institution. Mark the re-evaluated financial information as real-time financial data and mark the customer's financial information from other financial institutions as historical financial data.

[0008] S2: Represent historical financial data in the form of a knowledge graph. Determine the correlation between real-time financial data and historical financial data based on the shortest path distance from entities in real-time financial data to entities in historical financial data. Determine the timeliness of historical financial data based on the timestamps of historical financial data.

[0009] S3: Conduct a comprehensive analysis of the correlation between real-time financial data and historical financial data, as well as the timeliness of historical financial data, to determine the data that can be used as reference data for the current financial institution, mark it as storable data, and store the storable data in the customer's financial file in the current financial institution.

[0010] S4: Merge customer storable data with real-time financial data, determine the fluctuation of financial amounts based on the standard deviation and average value of financial amounts in the merged financial data, and determine the frequency information and speed information of transactions in the merged financial data based on the standard values ​​of the optimal transaction frequency range and single transaction rate in financial institutions.

[0011] S5: Comprehensively analyze the fluctuation of financial amounts within a unit of time, the frequency of transactions, and the speed of a single transaction. Based on the time length of the merged financial data, determine the number of comprehensive analysis results, generate several sets of comprehensive analysis results for the merged financial data, and determine the financial data that needs to be stored separately in the merged financial data.

[0012] In a preferred embodiment, step 2, determining the correlation between real-time financial data and historical financial data, and the timeliness of historical financial data, includes:

[0013] Construct a knowledge graph of current financial institutions, model historical and real-time financial data as entities in the knowledge graph, connect them through relationships, and ensure that each entity has corresponding attributes;

[0014] Obtain the shortest path from the entity in historical financial data to the entity in real-time financial data, and set the threshold for the shortest path from the entity in historical financial data to the entity in real-time financial data.

[0015] The correlation coefficient of the spectral data is calculated using a formula.

[0016] Obtain the timestamps of historical financial data, and based on the timestamps of each customer's historical financial data, obtain the length of the time interval between each customer's historical financial data and the customer's transfer to the current financial institution;

[0017] Set a threshold for the time interval between each customer's historical financial data and the customer's transfer to the current financial institution;

[0018] The timeliness coefficient of the data is calculated using a formula.

[0019] In a preferred embodiment, step 3, determining the data that can be used as reference data for the current financial institution, includes:

[0020] Based on the correlation between historical financial data and real-time financial data and the timeliness of historical financial data, the correlation coefficient and timeliness coefficient of the graph data are obtained.

[0021] The correlation coefficient and timeliness coefficient of the graph data are combined to generate the financial data storage evaluation coefficient;

[0022] The correlation coefficient and timeliness coefficient of map data are positively correlated with the evaluation coefficient of financial data storage.

[0023] In a preferred embodiment, in step 3, based on the financial data storage evaluation coefficient, data that can serve as reference data for the current financial institution is marked as storable data, including:

[0024] Compare the evaluation coefficient of financial data storage with the threshold of financial data storage evaluation coefficient for historical financial data;

[0025] If the financial data storage evaluation coefficient is greater than the financial data storage evaluation coefficient threshold, then the historical financial data will be marked as storable data and stored in the customer's financial file.

[0026] If the financial data storage evaluation coefficient is less than the financial data storage evaluation coefficient threshold, then historical financial data will not be used as long-term reference information for current financial institutions to evaluate customer financial information.

[0027] In a preferred embodiment, step 4 involves determining the fluctuations in financial amounts, transaction frequency information, and speed information of individual transactions within the merged financial data, including:

[0028] Collect the financial amount of customer transactions in financial institutions within a unit time t. Based on the financial amount of customer transactions in financial institutions within a unit time t, calculate the standard deviation and average of the financial amount of each customer transaction. Calculate the coefficient of variation of the financial data using a formula.

[0029] Obtain the optimal trading frequency range of customer financial data within a unit time t, and obtain the trading frequency of customer financial data within a unit time t. Calculate the financial data range anomaly coefficient using a formula.

[0030] Obtain the standard value of a customer's single transaction rate at a financial institution, and obtain the transaction speed of each transaction within a unit time t. Calculate the financial data speed offset coefficient using a formula.

[0031] In a preferred embodiment, step 5 involves a comprehensive analysis of the fluctuations in the amount of money involved per unit time, the frequency of transactions, and the speed of individual transactions, including:

[0032] Based on the fluctuations in the financial amount of customer transactions, transaction frequency information, and transaction speed information, we obtain the financial data variation coefficient, financial data range anomaly coefficient, and financial data rate deviation coefficient.

[0033] The coefficient of variation of financial data, the coefficient of anomaly of the range of financial data, and the coefficient of rate deviation of financial data are combined to generate the classification and storage evaluation coefficient.

[0034] The coefficient of variation of financial data, the coefficient of anomaly of financial data range, and the coefficient of rate deviation of financial data are positively correlated with the classification and storage evaluation coefficient.

[0035] In a preferred embodiment, in step 5, based on the time length of the merged financial data, the number of comprehensive analysis results is determined, several sets of comprehensive analysis results for the merged financial data are generated, and the financial data that needs to be stored separately from the merged financial data is determined, including:

[0036] Obtain a set of classification and storage evaluation coefficients for customer financial data, and compare the classification and storage evaluation coefficient threshold with the classification and storage evaluation coefficients in the set of classification and storage evaluation coefficients.

[0037] If the classification storage evaluation coefficient is greater than the classification storage evaluation coefficient threshold, then the unit time period t corresponding to this classification storage evaluation coefficient is marked to generate a time period set;

[0038] If the classification storage evaluation coefficient is less than the classification storage evaluation coefficient threshold, then no marking is performed;

[0039] Obtain the financial data of customers within each unit time t in the time period set, and mark it as separately stored financial data.

[0040] The technical effects and advantages of this invention are as follows:

[0041] 1. This invention comprehensively assesses which historical financial data can be stored in customers' financial files as long-term reference information for current financial institutions by considering the correlation between historical financial data and real-time financial data, as well as the time interval between historical financial data and real-time financial data. This helps financial institutions to filter historical financial data from other financial institutions and retain meaningful financial information for customers.

[0042] 2. This invention uses a comprehensive analysis of the financial amount information, frequency information, and speed information of customer transactions to determine which financial information contains risky information. This risky financial information is stored separately, which helps financial institutions to more easily identify potential risks. By storing it separately, potential risky financial information can be located more accurately. Attached Figure Description

[0043] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0044] Figure 1 This is a flowchart illustrating the steps of a method for classifying and storing information technology data according to the present invention. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] Example 1

[0047] Figure 1 This invention provides a method for classifying and storing information technology data, which includes the following steps:

[0048] S1: Transfer the customer's financial information from other financial institutions to the current financial institution, and re-evaluate the customer's financial information at the current financial institution. Mark the re-evaluated financial information as real-time financial data and mark the customer's financial information from other financial institutions as historical financial data.

[0049] S2: Represent historical financial data in the form of a knowledge graph. Determine the correlation between real-time financial data and historical financial data based on the shortest path distance from entities in real-time financial data to entities in historical financial data. Determine the timeliness of historical financial data based on the timestamps of historical financial data.

[0050] S3: Conduct a comprehensive analysis of the correlation between real-time financial data and historical financial data, as well as the timeliness of historical financial data, to determine the data that can be used as reference data for the current financial institution, mark it as storable data, and store the storable data in the customer's financial file in the current financial institution.

[0051] S4: Merge customer storable data with real-time financial data, determine the fluctuation of financial amounts based on the standard deviation and average value of financial amounts in the merged financial data, and determine the frequency information and speed information of transactions in the merged financial data based on the standard values ​​of the optimal transaction frequency range and single transaction rate in financial institutions.

[0052] S5: Comprehensively analyze the fluctuation of financial amounts within a unit of time, the frequency of transactions, and the speed of a single transaction. Based on the time length of the merged financial data, determine the number of comprehensive analysis results, generate several sets of comprehensive analysis results for the merged financial data, and determine the financial data that needs to be stored separately in the merged financial data.

[0053] In step 1, due to differences in the business types of financial institutions, the competitiveness of different businesses, and the customer's own reasons such as changes in the customer's finances and changes in the degree of trust in the financial institution, the customer may frequently switch between different financial institutions. When a customer moves from one financial institution to another, different financial institutions may use different financial assessment indicators and models, resulting in different ratings of the customer's financial information by different financial institutions under the same financial information. Therefore, the customer needs to reassess the customer's financial information in the current financial institution.

[0054] It should be noted that financial institutions typically provide different financial services based on the rating of a customer's financial information. These different financial services include offering more competitive loan interest rates, credit card limits, loan approval probabilities, and lower insurance premiums, etc., based on the customer's financial information rating.

[0055] Customer financial information from other financial institutions is labeled as historical financial data, while financial information re-examined within the current financial institution is labeled as real-time financial data.

[0056] It should be noted that a customer's financial information at other financial institutions is their previous financial information. The difference in financial information is not only reflected in the fact that different financial institutions may need different financial information, but also in the fact that, in terms of time, the customer's financial information at other financial institutions is lagging behind the financial information re-examined at the current financial institution. The re-examined financial information represents the financial information generated after the customer transferred to the current financial institution.

[0057] In step 2, depending on the financial institution, the business focus of the financial institution is different. The types of financial information that the current financial institution re-evaluates are different from those of other financial institutions. Therefore, when constructing a knowledge graph, the current financial institution chooses to focus on different detection indicators based on its own characteristics, business direction and customer needs, thereby forming a customized knowledge graph.

[0058] Construct a knowledge graph of current financial institutions, model historical and real-time financial data as entities in the knowledge graph, connect them through relationships, and ensure that each entity has corresponding attributes, such as transaction amount and daily revenue. Use graph query language or graph algorithm to calculate the shortest path from entities in historical financial data to entities in real-time financial data.

[0059] It should be noted that constructing a knowledge graph for current financial institutions is beneficial for modeling all real-time financial data as entities in the knowledge graph. However, historical financial data may not necessarily be able to be connected to entities in real-time financial data. This is because the data formats of historical and real-time financial data are inconsistent, and different financial institutions focus on different levels of richness of financial data. In other words, the types of historical financial data and real-time financial data may be different.

[0060] Based on the shortest path from entities in historical financial data to entities in real-time financial data, the correlation between historical and real-time financial data can be approximated. In the knowledge graph, the shorter the shortest path distance from entities in historical financial data to entities in real-time financial data, the higher the correlation between historical and real-time financial data, and the greater the possibility that historical financial data will be stored in existing financial institutions. Conversely, the longer the shortest path distance from entities in historical financial data to entities in real-time financial data, the lower the correlation between historical and real-time financial data, and the less likely that historical financial data will be stored in existing financial institutions.

[0061] It should be noted that storing historical financial data in existing financial institutions is for the purpose of building more complete and accurate customer financial profiles and classifying historical financial data. For historical financial data with low relevance, existing financial institutions choose not to store it, thereby reducing data management costs and avoiding a decrease in attention to truly important data.

[0062] The logic for obtaining the correlation coefficient of graph data is as follows: obtain the shortest path from an entity in historical financial data to an entity in real-time financial data, and mark it as: Set a shortest path threshold from entities in historical financial data to entities in real-time financial data, and mark the shortest path threshold as: ;

[0063] The correlation coefficient of the spectral data is calculated using the following formula: ;in, The correlation coefficient of the spectral data;

[0064] It should be noted that the higher the correlation coefficient of the graph data, the higher the correlation between historical financial data and real-time financial data, indicating that historical financial data is more likely to be used as a long-term financial assessment indicator and that historical financial data is more likely to be stored in existing financial institutions.

[0065] Based on the timestamps of the customer's historical financial data, the timeliness of the historical financial data is determined. By analyzing the timeliness of the historical financial data, it is determined whether the historical financial data has substantial reference value. The greater the timeliness of the historical financial data, the more recently generated the historical financial data is, and the more recently generated the customer's historical financial data has substantial reference value.

[0066] The logic for obtaining the data timeliness coefficient is as follows: Obtain the timestamps of historical financial data; based on the timestamps of each customer's historical financial data, obtain the time interval between each historical financial data entry and the customer's transfer to the current financial institution; and mark the time interval length as: n = 1, 2, 3...N, where N is a positive integer, and n represents the number of the time interval between each historical financial data of the customer and the customer's transfer to the current financial institution;

[0067] Set a threshold for the time interval between each customer's historical financial data and their transfer to the current financial institution, and mark the threshold as: The timeliness coefficient of the data is calculated using the following formula: ;in, This is the data timeliness coefficient.

[0068] As the formula shows, the larger the data timeliness coefficient, the shorter the time interval between historical financial data and real-time financial data, indicating that the customer's historical financial data has more substantial reference value.

[0069] In step 3, based on the correlation between historical financial data and real-time financial data, and the timeliness of historical financial data, the correlation coefficient and timeliness coefficient of the graph data are comprehensively analyzed to establish an evaluation model and generate a financial data storage evaluation coefficient. The formula for calculating the financial data storage evaluation coefficient is as follows: ;in, This is a financial data storage evaluation coefficient for historical financial data. This refers to the proportional coefficients of the correlation coefficient and the timeliness coefficient of the spectral data. Greater than 0.

[0070] As can be seen from the formula, the larger the correlation coefficient and timeliness coefficient of the graph data, the larger the financial data storage evaluation coefficient, indicating that existing financial institutions believe that historical financial data is more likely to become long-term reference information for customers' financial information and will store historical financial data in customers' financial files.

[0071] The evaluation coefficient of historical financial data storage is compared with the threshold of financial data storage evaluation coefficient. If the evaluation coefficient of financial data storage is greater than the threshold, the historical financial data is marked as storable and stored in the customer's financial file. If the evaluation coefficient of financial data storage is less than the threshold, the historical financial data is not used as long-term reference information for the current financial institution to evaluate the customer's financial information.

[0072] This embodiment comprehensively assesses which historical financial data can be stored in customers' financial files as long-term reference information by considering the correlation between historical and real-time financial data, as well as the time interval between historical and real-time financial data. This helps financial institutions screen large amounts of customers' historical financial data and retain meaningful financial information from customers.

[0073] Example 2

[0074] Through the above embodiments, storable data is combined with real-time financial data to form the customer's financial information at the current financial institution. By comprehensively analyzing the financial amount, transaction frequency, and transaction speed in the financial information, potential risk information in the customer's financial information is identified.

[0075] In step 4, by combining the customer's storable data and real-time financial data, and through the customer's overall financial information, we analyze the financial information with potential risks in the customer's financial information, classify the financial information with potential risks with other financial information, and store the financial information with potential risks separately.

[0076] It should be noted that storing potentially risky financial data separately helps financial institutions to rate customer financial information, making it easier to deal with the uncertainty and volatility of customer transaction financial information, and thus gain a more comprehensive understanding of customer behavior and preferences.

[0077] The logic for obtaining the coefficient of variation of financial data is as follows: collect the financial amount of customer transactions within a unit of time t, and label the financial amount of each customer transaction as follows: Where m = 1, 2, 3...M, M is a positive integer, and m is the number of the customer's transaction behavior in the financial institution;

[0078] It should be noted that after a customer transfers from another financial institution to their current financial institution, the current financial institution will conduct an assessment of the customer for a period of time. This assessment typically involves evaluating the customer's liquidity and payment ability based on their transaction behavior in their current financial institution account, including deposits, withdrawals, transfers, and spending.

[0079] Based on the amount of financial transactions made by customers at financial institutions within a unit of time t, calculate the standard deviation and average of the financial amount for each customer transaction, and label the standard deviation and average of the financial amount as follows: and The calculation formula is: , ;

[0080] The formula for calculating the coefficient of variation of financial data is as follows: ;in, For a unit of time The coefficient of variation of financial data within the country The coefficient of variation represents the amount of financial transactions made by a customer within a unit of time t.

[0081] As the formula shows, the larger the coefficient of variation of financial data, the greater the fluctuation of the financial amount in the transaction, indicating that the customer's transaction behavior is uncertain and the financial amount of each transaction is highly volatile. Therefore, the customer's financial data should be stored separately to facilitate the analysis of abnormal transaction behavior.

[0082] The logic for obtaining the anomaly coefficient of financial data range is as follows: obtain the optimal transaction frequency range of customer financial data within a unit time t, and mark the optimal transaction frequency range as: , This represents the minimum number of transactions a customer can make within a unit of time t. The highest number of transactions by a customer within a unit of time t;

[0083] It should be noted that the optimal trading frequency range should be set according to the financial institution's business type and specific rules. If the number of transactions a customer makes within a unit of time t is less than the range threshold, it indicates that the customer's financial account is an inactive account. If the number of transactions within a unit of time t is greater than the range threshold, it may indicate that the customer's financial account has abnormal high-frequency trading activities.

[0084] Obtain the transaction frequency of customer financial data within a unit time t, and mark the transaction frequency of customer financial data as: ;

[0085] The formula for calculating the anomaly coefficient of financial data range is as follows: ;in, This is the anomaly coefficient for the range of financial data;

[0086] As the formula shows, the larger the anomaly coefficient of the financial data range, the more likely the customer's financial account is not a normal account. Therefore, the customer's financial data may be stored separately to facilitate the analysis of abnormal transaction behavior.

[0087] The logic for obtaining the financial data rate offset coefficient is as follows: obtain the standard value of a customer's single transaction rate at a financial institution, and mark the standard value as: ;

[0088] It should be noted that the standard transaction speed set by financial institutions is to ensure market fairness. Transaction speed that is too fast allows them to obtain key market information and execute trades before other traders, which may cause the market to become unbalanced. Transaction speed that is too slow may cause clients to miss favorable market opportunities. Financial institutions should set a standard transaction speed based on the general trading speed of the public.

[0089] Obtain the transaction speed of each transaction made by the customer within a unit time t, and label the transaction speed as: ;

[0090] The formula for calculating the financial data rate offset coefficient is as follows: ;in, This is the financial data rate offset coefficient;

[0091] As the formula shows, the larger the financial data rate deviation coefficient, the greater the difference between the speed at which a customer completes each transaction within a unit time t and the standard value. This indicates that the customer may be engaging in high-frequency trading or other security issues that cause transaction delays. Therefore, it is necessary to store the customer's financial data separately to facilitate the analysis of abnormal trading behavior.

[0092] In step 4, the customer's storable data and real-time financial data are combined and analyzed. Based on the fluctuations in the financial amount of customer transactions, transaction frequency information, and transaction speed information, the coefficient of variation of financial data, the coefficient of anomaly of financial data range, and the coefficient of rate deviation of financial data are comprehensively analyzed to establish a data analysis model and generate a classification storage evaluation coefficient. The calculation formula for the classification storage evaluation coefficient is as follows: ,in, The evaluation coefficient for customer's categorized storage within a unit of time t. These are respectively represented as the coefficient of variation of financial data, the coefficient of anomaly in the range of financial data, and the coefficient of rate deviation of financial data. These represent the proportional coefficients for the coefficient of variation of financial data, the coefficient of anomaly in the range of financial data, and the coefficient of rate deviation of financial data, respectively. Greater than 0.

[0093] As shown in the formula, the larger the financial data variation coefficient, financial data range anomaly coefficient, and financial data rate offset coefficient, the larger the classification storage evaluation coefficient, indicating a higher probability of abnormal customer transaction behavior within a unit of time t, and a higher probability of customer financial data being stored separately within a unit of time t. Conversely, the smaller the financial data variation coefficient, financial data range anomaly coefficient, and financial data rate offset coefficient, the smaller the classification storage evaluation coefficient, indicating a lower probability of abnormal customer transaction behavior within a unit of time t, and a lower probability of customer financial data being stored separately within a unit of time t. The less likely it is that financial data within the organization will be stored separately.

[0094] It should be noted that the unit time t is a relatively short period of time. The specific range should be determined based on the business type of the financial institution, the financial institution itself, and the length of time for storing customer data and real-time financial data. This is because the greater the possibility of abnormal transaction behavior by the customer within the unit time t, the greater the potential risk of the customer's financial information within the unit time t. Therefore, it is necessary to store the financial information within the unit time t separately.

[0095] Based on the time frame of the merged financial data, the number of comprehensive analysis results is determined, and a set of classification and storage evaluation coefficients for customer financial data is obtained. This set of classification and storage evaluation coefficients is then labeled as follows: , where z = 1, 2, 3... Z, Z is a positive integer, and z represents the number of the classification storage evaluation coefficient for each unit time t in the customer classification storage evaluation coefficient set;

[0096] Set the classification storage evaluation coefficient threshold, and mark the classification storage evaluation coefficient threshold as: The classification storage evaluation coefficient threshold is compared with the classification storage evaluation coefficients in the classification storage evaluation coefficient set. If the classification storage evaluation coefficient is greater than the classification storage evaluation coefficient threshold, the unit time period t corresponding to this classification storage evaluation coefficient is marked, generating a time period set, and marked as follows: Where l = 1, 2, 3...L, L is a positive integer, and l is the number of the time period in the time period set. If the classification storage evaluation coefficient is less than the classification storage evaluation coefficient threshold, it will not be marked.

[0097] Obtain customer financial data for each unit of time t within the time period set and mark it as separately stored financial data. Separately stored financial data indicates that the portion of customer financial data needs to be stored separately. Through professional analysis of the separately stored portion, it is possible to understand customer behavior and more effectively manage customer financial information ratings at financial institutions.

[0098] This embodiment uses a comprehensive analysis of customer transaction financial amount information, transaction frequency information, and transaction speed information to determine which risky financial information exists in the customer's financial information, and stores the risky financial information separately. This helps financial institutions to more easily identify potential risks, and by storing them separately, they can more accurately locate the financial information with potential risks.

[0099] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0100] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0101] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0102] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0103] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0104] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0105] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A classification storage method for information technology data, characterized in that: The following steps are involved: S1: Transfer the customer's financial information from other financial institutions to the current financial institution, and re-evaluate the customer's financial information at the current financial institution, mark the re-evaluated financial information as real-time financial data, and mark the customer's financial information from other financial institutions as historical financial data; S2: Represent historical financial data in the form of a knowledge graph, determine the correlation between real-time financial data and historical financial data based on the shortest path distance from the entity of real-time financial data to the entity of historical financial data, and determine the timeliness of historical financial data based on the timestamp of the historical financial data; S3: Comprehensively analyze the correlation between real-time financial data and historical financial data and the timeliness of historical financial data, determine the data that can be used as reference data for the current financial institution, mark it as storable data, and store the storable data in the financial file of the customer at the current financial institution; S4: Merge the customer storable data with the real-time financial data, determine the fluctuation of the financial amount based on the standard deviation and average value of the financial amount in the merged financial data, and determine the frequency information of the transaction and the speed information of the single transaction in the merged financial data based on the optimal transaction frequency range and the standard value of the single transaction rate in the financial institution; S5: Comprehensively analyze the fluctuation of financial amounts within a unit time, the frequency information of transactions, and the speed information of single transactions, determine the number of comprehensive analysis results based on the time length of the merged financial data, generate a set of results of several comprehensive analyses of the merged financial data, and determine the financial data that needs to be stored separately in the merged financial data.

2. A classification storage method for information technology data according to claim 1, characterized in that: In step 2, the relevance of real-time financial data to historical financial data and the timeliness of historical financial data are determined, including: Build a knowledge graph of current financial institutions, model historical financial data and real-time financial data as entities in the knowledge graph, connect them through relationships, and ensure that each entity has corresponding attributes; Obtaining the shortest path from an entity of historical financial data to an entity of real-time financial data, and setting a threshold for the shortest path from an entity of historical financial data to an entity of real-time financial data; Calculate the correlation coefficient of the atlas data through the formula; Obtain the timestamp of historical financial data, and obtain the time interval between each historical financial data of the customer and the transfer of each historical financial data to the current financial institution according to the timestamp of each historical financial data of the customer; Set the threshold for the length of the time interval between each historical financial data of the customer and the transfer of the customer to the current financial institution; Calculate the data timeliness coefficient through the formula.

3. A classification storage method for information technology data according to claim 2, characterized in that: In step 3, determine the reference data that can be used as the current financial institution, including: According to the relevance of historical financial data to real-time financial data and the timeliness of historical financial data, the graph data relevance coefficient and data timeliness coefficient are obtained; The graph data correlation coefficient and the data timeliness coefficient are combined to generate the financial data storage evaluation coefficient; The graph data correlation coefficient and data timeliness coefficient are positively correlated with the financial data storage evaluation coefficient.

4. A classification storage method for information technology data according to claim 3, characterized in that: In step 3, based on the financial data storage evaluation coefficient, the reference data of the current financial institution is marked as storable data, including: comparing a financial data storage evaluation coefficient of the historical financial data with a financial data storage evaluation coefficient threshold; If the financial data storage evaluation coefficient is greater than the financial data storage evaluation coefficient threshold, the historical financial data is marked as storable data, and the historical financial data is stored in the customer's financial file; If the financial data storage evaluation coefficient is less than the financial data storage evaluation coefficient threshold, the historical financial data will not be used as long-term reference information for the current financial institution to evaluate customer financial information.

5. A classification storage method for information technology data according to claim 4, characterized in that: In step 4, the fluctuation of financial amounts, the frequency information of transactions and the speed information of single transactions in the merged financial data are determined, including: Collect the financial amount of the customer's transaction behavior in the financial institution within the unit time t, calculate the standard deviation and average value of the financial amount in each customer transaction behavior based on the financial amount of the customer's transaction behavior in the financial institution within the unit time t, and calculate the coefficient of variation of the financial data through the formula; Obtain the optimal transaction frequency range of the customer's financial data within the unit time t, and obtain the transaction frequency of the customer's financial data within the unit time t, and calculate the abnormal coefficient of the financial data range through the formula; Obtain the standard value of the customer's single transaction rate in a financial institution, and obtain the transaction speed of each transaction of the customer within unit time t, and calculate the financial data speed deviation coefficient through the formula.

6. A classification storage method for information technology data according to claim 5, characterized in that: In step 5, the fluctuation of financial amount per unit time, the frequency information of transactions and the speed information of single transactions are comprehensively analyzed, including: According to the fluctuation of the financial amount of the customer's transaction, the frequency information of the transaction and the speed information of the transaction, the coefficient of variation of the financial data, the coefficient of abnormality of the financial data range and the coefficient of deviation of the financial data rate are obtained; The financial data variation coefficient, the financial data range anomaly coefficient and the financial data rate deviation coefficient are jointly generated into a classification storage evaluation coefficient; The coefficient of variation of financial data, the coefficient of abnormality of financial data range and the coefficient of rate deviation of financial data are positively correlated with the classification storage evaluation coefficient.

7. A classification storage method for information technology data according to claim 6, characterized in that: In step 5, the number of comprehensive analysis results is determined according to the time length of the merged financial data, a result set of several comprehensive analyses of the merged financial data is generated, and the financial data that needs to be stored separately in the merged financial data is determined, including: Obtaining a classification storage evaluation coefficient set for the customer's financial data, and comparing the classification storage evaluation coefficient threshold with the classification storage evaluation coefficient in the classification storage evaluation coefficient set; If the classification storage evaluation coefficient is greater than the classification storage evaluation coefficient threshold, the unit time period t corresponding to the classification storage evaluation coefficient is marked to generate a time period set; If the classification storage evaluation coefficient is less than the classification storage evaluation coefficient threshold, no marking is performed; The financial data of the customer in each unit time t in the time period set is obtained and marked as separately stored financial data.

Images