Method, system, storage medium and device for acquiring financial service data discrimination

Abstract

The embodiment of the application discloses a method for obtaining the discrimination of financial service data, which comprises the following steps: obtaining a candidate financial service data set as a financial data sample, determining the first number of different values R1 of the financial data sample according to part of the data of the financial data sample, determining the second number of different values R2 of the financial data sample according to all the data of the financial data sample, and finally determining the discrimination of the financial service data according to the comparison relationship between the first number of different values R1 and the second number of different values R2, so that the related service data can be quickly inquired or retrieved according to the size of the discrimination of the data in the financial service data set.

Description

Technical Field

[0001] This invention relates to the field of digital data processing technology, and in particular to methods, systems, storage media, and devices for obtaining the distinguishability of financial business data. Background Technology

[0002] In recent years, with the rapid development of new technologies such as big data, cloud computing, blockchain, and artificial intelligence, these technologies have been deeply integrated with financial business, releasing the vitality of financial innovation and application potential. This has made the integration, sharing, and openness of financial industry data a trend. However, the complexity and large volume of financial industry data can easily lead to the dispersion and chaos of financial sample data, resulting in problems such as low retrieval efficiency, incomplete data retrieval, and retrieval errors. Summary of the Invention

[0003] Therefore, it is necessary to propose a method for obtaining the distinguishability of financial business data in response to the above problems.

[0004] A method for obtaining the distinguishability of financial business data, the method comprising the following steps:

[0005] Acquire candidate financial business datasets to create financial data samples;

[0006] The number of distinct values ​​R1 of the financial data sample is determined based on a portion of the data from the financial data sample; the number of distinct values ​​R2 of the financial data sample is determined based on all the data from the financial data sample; and the discriminative power of the financial business data is determined based on the comparison between the number of distinct values ​​R1 and the number of distinct values ​​R2.

[0007] In the above scheme, obtaining candidate financial business datasets as financial data samples specifically includes: obtaining candidate financial business datasets from the database as financial data samples.

[0008] In the above scheme, obtaining candidate financial business datasets from the database as financial data samples includes periodically updating the database or manually adding or deleting database content.

[0009] In the above scheme, determining the number R1 of the first distinct values ​​of the financial data sample based on a portion of the financial data sample specifically includes:

[0010] Half of the financial data sample is selected to determine the first distinct value R1 of the financial data sample.

[0011] In the above scheme, the step of selecting half of the financial data sample to determine the first distinct value R1 of the financial data sample specifically includes:

[0012] The financial data sample is obtained from the interval [min-id, min-id+(max-id-min-id) / 2], where min-id is the identifier number of the smallest data and max-id is the identifier number of the largest data.

[0013] In the above scheme, determining the number R1 of the first distinct values ​​of the financial data sample based on a portion of the financial data sample specifically includes:

[0014] Based on the screening criteria, the effective data in the financial data sample is obtained as part of the data to determine the first distinct value R1 of the financial data sample.

[0015] In the above scheme, determining the number of second distinct values ​​R2 of the financial data sample based on all the data of the financial data sample specifically includes:

[0016] All data are obtained from the interval [min-id, max-id], where min-id is the identifier number of the minimum data and max-id is the identifier number of the maximum data.

[0017] In the above scheme, determining the discriminative power of financial business data based on the comparison relationship between the first number of different values ​​R1 and the second number of different values ​​R2 further includes: determining the data slope X of the financial data sample, where X = R2 / R1.

[0018] In the above scheme, determining the data discriminability based on the comparison relationship between the first number of different values ​​R1 and the second number of different values ​​R2 specifically includes:

[0019] When the data slope X ≤ a fixed value, the number of the second distinct values ​​R2 is used as the discrimination metric for the financial data sample;

[0020] When the data slope X > a fixed value, the number of the second distinct values ​​R2 is amplified according to the sampling ratio, and the amplified number of the second distinct values ​​R2 is used as the discrimination of the financial data sample.

[0021] In the above scheme, determining the data discrimination based on the comparison relationship between the first number of different values ​​R1 and the second number of different values ​​R2 further includes: automatically obtaining the first number of different values ​​R1 and the second number of different values ​​R2 using functions in the database.

[0022] This application also proposes a system for obtaining the distinguishability of financial business data, characterized in that the system includes: a financial data sample acquisition unit, a first distinct value acquisition unit, a second distinct value acquisition unit, and an output unit;

[0023] The financial data sample acquisition unit is used to acquire candidate financial business datasets as financial data samples;

[0024] The first distinct value acquisition unit is used to determine the first distinct value of the financial data sample based on a portion of the data of the financial data sample.

[0025] The second distinct value acquisition unit is used to determine the second distinct value of the financial data sample based on all the data of the financial data sample;

[0026] The output unit is used to determine the distinguishability of financial business data based on the comparison relationship between the first number of different values ​​and the second number of different values.

[0027] This application also proposes a readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps:

[0028] Acquire candidate financial business datasets to create financial data samples;

[0029] Determine the first distinct value R1 of the financial data sample based on partial data from the financial data sample;

[0030] Determine the second distinct value R2 of the financial data sample based on all the data in the financial data sample;

[0031] The discriminability of financial business data is determined based on the comparison relationship between the first number of different values ​​R1 and the second number of different values ​​R2.

[0032] This application also proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps: acquiring candidate financial business datasets and converting them into financial data samples;

[0033] Determine the first distinct value R1 of the financial data sample based on partial data from the financial data sample;

[0034] Determine the second distinct value R2 of the financial data sample based on all the data in the financial data sample;

[0035] The discriminability of financial business data is determined based on the comparison relationship between the first number of different values ​​R1 and the second number of different values ​​R2.

[0036] The embodiments of the present invention have the following beneficial effects: Candidate financial business datasets are obtained as financial data samples, and the first number of distinct values ​​R1 of the financial data samples is determined based on a portion of the data in the financial data samples. Then, the second number of distinct values ​​R2 of the financial data samples is determined based on all the data in the financial data samples. Finally, the discriminative power of the financial business data is determined based on the comparison relationship between the first number of distinct values ​​R1 and the second number of distinct values ​​R2. After obtaining the discriminative power of the financial business data, relevant business data can be quickly queried or retrieved based on the magnitude of the discriminative power in the financial business data set.

[0037] This method incorporates the concept of data discriminability, which comprehensively considers the meaning or information represented by the financial business data itself, thereby more accurately and quickly identifying and retrieving financial business data, which helps improve the efficiency of financial business data processing. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] in:

[0040] Figure 1 This is a schematic diagram of a method for obtaining the distinguishability of financial business data in one embodiment;

[0041] Figure 2 This is a flowchart illustrating the process of determining the distinguishability of financial business data in one embodiment;

[0042] Figure 3 This is a flowchart for determining the distinguishability of financial business data in this application. Detailed Implementation

[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0044] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0045] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention may be practiced without one or more of these details.

[0046] In other examples, some technical features well-known in the art have not been described to avoid confusion with the present invention. It should be understood that the present invention can be implemented in different forms and should not be construed as being limited to the embodiments set forth herein.

[0047] Instead, providing these embodiments will make the disclosure thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0048] The terminology used herein is intended only to describe particular embodiments and is not intended to limit the invention. When used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms, and the terms “comprising” and / or “including” should be understood unless the context clearly indicates otherwise.

[0049] When used in this specification, the presence of the stated features, integers, steps, operations, elements, and / or components is confirmed, but the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups is not excluded. As used herein, the term "and / or" includes any and all combinations of the associated listed items.

[0050] To fully understand the present invention, a detailed structure will be presented in the following description in order to illustrate the technical solution proposed by the present invention. Optional embodiments of the present invention are described in detail below. However, in addition to these detailed descriptions, the present invention may have other embodiments.

[0051] The financial industry has always attached great importance to the development of big data technology. The rapid construction and implementation of "financial cloud" has laid the foundation for the application of financial big data. The integration and application of financial data with other cross-domain data is constantly being strengthened. Artificial intelligence is becoming a new direction for the application of financial big data. The integration, sharing and opening of financial industry data is becoming a trend, bringing new development opportunities and huge impetus to the financial industry.

[0052] Existing financial databases are mainly divided into TP and AP, which are transactional and analytical databases, respectively. From a global market perspective, the growth of transactional databases has slowed down in the past two years and has been surpassed by analytical databases. This is because transactional databases are related to transaction volume, which is related to the total population and the slowdown in economic growth.

[0053] Analytical databases are involved in a wide range of business applications. Whether it's intelligent recommendation, customer profiling, or various analytical queries, they all require analytical data warehouses as the underlying support. In particular, in the past two years, domestic databases have risen against the trend, and the demand is strong.

[0054] However, the sheer size and complexity of financial data in databases, including its diverse formats, categories, and sources, lead to inefficient processing and analysis. Accurately locating relevant financial data is beneficial for obtaining valuable financial information, thereby effectively improving the efficiency of financial data processing and analysis.

[0055] Therefore, this application proposes a method for quickly and accurately finding financial business data based on the discriminative power of financial data.

[0056] like Figure 1 As shown, in one embodiment, a method for obtaining the distinguishability of financial business data is provided. This method includes steps S101 to S106, detailed below:

[0057] S101. Obtain candidate financial business datasets and combine them into financial data samples;

[0058] In some embodiments, candidate financial business datasets can be obtained from existing financial databases as financial data samples, such as: Wind, Bloomberg, Factset, Capital IQ, Thomson Reuters, iFind, CSMAR, and other general-purpose or specialized financial databases.

[0059] Specifically, different databases have different advantages and disadvantages, and cover different professional fields. The specific fields covered by the existing financial databases mentioned above are as follows:

[0060] The Wind database mainly covers areas such as stocks, bonds, funds, foreign exchange, financial derivatives, commodities, macroeconomics, and financial news.

[0061] The Bloomberg database primarily covers various asset classes, including fixed income, stocks, foreign exchange, commodities, and derivatives.

[0062] The Factset database primarily covers areas such as stock selection, stock research, stock order management, stock trading execution, and stock trading execution reports.

[0063] The Capital IQ database primarily covers: detailed information on private equity from both publicly and privately-owned companies worldwide;

[0064] The Thomson Reuters database primarily covers areas such as law, tax and accounting, risk and compliance, and global trade management.

[0065] The iFind database mainly covers financial products such as stocks, bonds, funds, futures, indices, and foreign exchange. This financial database can provide the latest and most comprehensive market information, news reports, and basic data.

[0066] The CSMAR database mainly covers 18 major categories, including factor research, personal characteristics, green economy, stocks, companies, overseas, information, funds, bonds, industries, economy, and commodity futures.

[0067] Luobo Investment Research Database mainly covers data areas such as the stock market and securities.

[0068] Preferably, the aforementioned existing financial databases are all capable of periodically updating the financial data they contain or manually adding or removing financial database content.

[0069] In some embodiments, the aforementioned periodic updating of financial data contained in the financial database or manual addition or subtraction of financial database content specifically includes:

[0070] 1. The financial data in the financial database is updated by having relevant staff manually set the update cycle based on the quality of the data in actual applications;

[0071] 2. Set up timed automatic updates of financial data in the financial database.

[0072] S102. Determine the number of first distinct values ​​R1 of the financial data sample based on partial data from the financial data sample;

[0073] In some embodiments, half of the financial data sample can be selected to determine the first distinct value R1 of the financial data sample, specifically including:

[0074] Obtain financial data samples from the interval [min-id, min-id+(max-id-min-id) / 2], where min-id is the identifier of the smallest data in the financial data sample, and max-id is the identifier of the largest data in the financial data sample.

[0075] In some embodiments, the number of first distinct values ​​R1 of the financial data sample can be determined by obtaining valid financial data sample data from the financial data sample as part of the data based on the screening criteria.

[0076] The aforementioned different values ​​represent the distinguishing points between different financial data sets. The more distinguishing points there are, the more distinct values ​​exist between the financial data sets, and the greater the discriminative power of these data sets.

[0077] Preferably, relevant financial data samples can be selected based on financial screening indicators, such as financial data included in indicators such as current assets, fixed assets, liabilities, and profitability.

[0078] Preferably, relevant financial data samples can be selected based on financial screening parameters, such as: cash and cash equivalents, interest, inventory, dividends, prepayments, notes receivable, settlement reserves, and funds lent out. If the search codes for the above financial screening parameters in a certain financial database are 00, 01, 02, 03, 04, 05, 06, 07, and 08, then if you need to search for financial data information for a specific indicator, you only need to enter the corresponding search code. For example, if you need to search for relevant financial information on settlement reserves, you only need to enter the search code 07 to obtain the relevant settlement reserve information.

[0079] In some embodiments, a downsampling method may be used to obtain partial data from the full data of a financial data sample to determine the first number of distinct values ​​R1 of the financial data sample.

[0080] Specifically, downsampling, also known as under-sampling, is a non-heuristic method for removing large classes of data to balance the distribution of data classes. The underlying logic of this method is to balance the dataset and thus overcome the specificity of the algorithm.

[0081] Commonly used methods include Random Under-Sampling (RUS), which randomly excludes observations of large classes, and Focused Under-Sampling (FUS), which removes large class data at the boundary between two classes.

[0082] In some embodiments, a downsampling method is used to obtain partial data from all the data of the financial data sample to determine the first distinct value R1 of the financial data sample. Financial sample data can be obtained by linspace sampling or uniform sampling. This sampling method is simple and convenient and has a certain degree of randomness, which can ensure the randomness and universality of the obtained financial sample data.

[0083] S103. Determine the number of second distinct values ​​R2 of the financial data sample based on all the data in the financial data sample;

[0084] In some embodiments, all data is obtained from the interval [min-id, max-id] as a financial data sample, where min-id is the identifier number of the smallest data in the financial data sample and max-id is the identifier number of the largest data in the financial data sample. The specific data involved in the financial data sample can be located or queried according to the interval [min-id, max-id].

[0085] S104. Determine the discriminative power of financial business data based on the comparison relationship between the number of first distinct values ​​R1 and the number of second distinct values ​​R2.

[0086] Discrimination refers to the ability of data in the same database to be distinguished from other data by the differences in the numbers or parameters contained in the code. The number of differences equals the number of discriminations.

[0087] In some embodiments, such as gender information in user profiles in a financial database, this information is generally represented by 0, 1 or A, B to indicate two genders, so the distinguishability of this information is 2.

[0088] In some embodiments, such as an ID card number, which typically consists of 18 digits, it may include the province code, city code, district / county code, the holder's date of birth, the holder's local sequence number, and the holder's gender information. In this case, the distinguishability of such data information is 6.

[0089] In some embodiments, financial data comes in many forms, and is classified into four categories from low to high complexity: fundamental data, market data, analytical data, and alternative data. These four types of data contain more different values ​​of information indicators, categories, validity periods, and characteristics, resulting in greater differentiation.

[0090] Regarding fundamental data:

[0091] Fundamental data mainly consists of quarterly financial reports. These data are released later than the statistical period (for example, the first quarter financial report may not be released until mid-May). Unreleased data cannot be used. That is, if you are doing research on January to March, you cannot use financial report data released in May.

[0092] Fundamental data is frequently backfilled and reinstated: the former refers to missing values ​​in the data being assigned new values; the latter refers to the fact that data may be modified multiple times after it is published, and the data provider often simply overwrites the original values ​​with new values, even though these new values ​​obviously did not exist when the data was published.

[0093] Fundamental data is highly systematic and infrequent. Because it is easily accessible to the market, its value is limited and it needs to be analyzed in conjunction with other data types.

[0094] Regarding market data:

[0095] Market data includes all trading activity occurring on exchanges. It provides access to raw, unstructured data streams, such as FIX data and BWIC (bids wanted in competition) data. Analyzing this financial data can help predict trading events.

[0096] Unlike fundamental data, FIX data is extremely large, with approximately 10TB of data generated daily, making it very difficult to process. This necessitates the introduction of the concept of discriminative power to retrieve and process the data.

[0097] About Analytics:

[0098] Analytical data is derived from other data after processing. It typically comes from investment banks and research firms, and some companies sell statistical analyses of the data, such as sentiment extracted from news reports and social media platforms. The advantage of analytical data is that it has already been extracted; the disadvantages are that this data is expensive, its calculation logic may be flawed, and it is used by more than one person.

[0099] Regarding alternative data:

[0100] Alternative data can be generated from individuals (social media, web searches), business processes, and sensors (satellites, positioning, weather). For example, if asset data within a bank's system is highly sought after, including monitoring data on interbank lending, central bank reserves, fixed assets, and intangible assets, or if the bank's customer asset data is highly sought after, then the bank will monitor the customer's fixed assets, its own financial assets, other banks' financial assets, and intangible assets.

[0101] Specifically, for banks, assets are mainly divided into four parts:

[0102] 1) Interbank lending. Interbank lending is the short-term, temporary adjustment of funds between financial institutions. When a bank lends money to other banks, it is called lending out funds, which is an asset; when a bank borrows money from other banks, it is called borrowing funds, which is a liability.

[0103] 2) Central Bank Reserves. To ensure that commercial banks have sufficient solvency when faced with a sudden surge in deposit withdrawals, the central bank requires commercial banks to deposit a certain percentage of their cash reserves with the central bank. These reserves are called central bank reserves and are considered assets.

[0104] 3) Fixed assets. This refers to the fixed assets owned by the bank, such as real estate, office buildings, fixed equipment, and vehicles under the bank's name.

[0105] 4) Intangible assets. Such as a bank's brand, patents, copyrights, reputation, etc.

[0106] In addition, for bank customers, assets are mainly divided into four categories:

[0107] 1) Fixed assets. This refers to the fixed assets owned by the client, such as real estate, vehicles, goods, land, etc., under the name of an individual / company.

[0108] 2) Financial assets of the Bank. This includes deposit products, loan products, wealth management products, and asset management products purchased from the Bank. Collateral and pledged assets used for loans also belong to the customer's assets.

[0109] 3) Financial assets held by other banks. This includes deposit products, loan products, wealth management products, and asset management products purchased from other banks, as well as financial products purchased from this bank on behalf of third parties.

[0110] 4) Intangible assets. Such as the client's brand, patents, copyrights, reputation, etc.

[0111] Asset data can be used to analyze the economic strength of banks / bank customers, thereby enabling marketing ratings or risk ratings for banks / bank customers.

[0112] Using alternative data means dealing with unique and difficult-to-process data, but the more difficult the data is to store and manipulate, the more valuable it is.

[0113] It is evident that the discriminative power of data information is of great significance, especially in financial business. Based on the discriminative power of financial data, it is possible to further determine whether to use a fixed value or a discriminative power to quickly and accurately find the required financial indicators or financial data information in a large financial database. This greatly improves the efficiency of finding financial business data. Using discriminative power to find financial sample data can also reduce problems such as low retrieval efficiency, incomplete data retrieval, and retrieval errors.

[0114] In some embodiments, determining the discriminative power of financial business data further includes determining the data slope X of the financial data sample, wherein the specific expression for obtaining the data slope X is as follows:

[0115] X = R2 / R1;

[0116] In the above expression, R1 is the number of the first distinct values, and R2 is the number of the second distinct values.

[0117] like Figure 2 As shown, in some embodiments, the discriminative power of financial business data is determined based on the comparison relationship between the first number of distinct values ​​R1 and the second number of distinct values ​​R2, specifically including:

[0118] S401. When the data slope X ≤ a fixed value, use the number of second distinct values ​​R2 as the discrimination metric for the financial data sample.

[0119] S402. When the data slope X > a fixed value, the number of second distinct values ​​R2 is amplified according to the sampling ratio, and the amplified number of second distinct values ​​R2 is used as the discrimination of the financial data sample.

[0120] Preferably, the above fixed value can be 1.15.

[0121] In some embodiments, the value of the fixed value can be determined manually based on the specific use case and the application model of the data.

[0122] In some embodiments, in a financial database, if the data to be queried is a gender parameter, then all users in the financial database are divided into male users or female users based on this parameter.

[0123] At this point, the number of the first distinct values ​​R1 of the financial data sample is determined based on a portion of the data. Obviously, this number of distinct values ​​R1 is 2. Then, the number of the second distinct values ​​R2 of the financial data sample is determined based on all the data. Similarly, this number of distinct values ​​R2 is also 2.

[0124] If the slope of the financial data obtained by the ratio of the first number of distinct values ​​R1 and the second number of distinct values ​​R2 is 1, and it is determined to be less than a fixed value (in this application scenario, the fixed value is generally set to 1.15), then there is no need to amplify the size of the number of distinct values ​​according to the sampling ratio. The second number of distinct values ​​R2 can be directly used as the discrimination of the financial data sample.

[0125] If the number of second distinct values, R², is not used as the discriminant of the financial data sample, but instead the number of second distinct values, 2, is amplified according to the sampling ratio (e.g., if the sampling ratio is 1:100, then it is amplified by 100 times), then the final number of second distinct values, i.e., the discriminant of the financial data sample, is 200. This is obviously incorrect, because in fact, there are only 2 distinct values ​​for the user's gender in the entire financial database, i.e., the discriminant of the user's gender parameter is also 2.

[0126] Preferably, the number of the first distinct values ​​R1 and the number of the second distinct values ​​R2 can be automatically obtained using functions in the database, such as the count function.

[0127] In summary, after obtaining the distinguishability of financial business data, relevant business data can be quickly queried or retrieved based on the magnitude of the distinguishability of data in the financial business data set. This method combines the concept of data distinguishability and can comprehensively consider the meaning or information represented by the financial business data itself, thereby more accurately and quickly identifying and finding financial business data, which is conducive to improving the efficiency of financial business data processing.

[0128] This application also proposes a system for obtaining the distinguishability of financial business data, the system comprising: a financial data sample acquisition unit, a first distinct value acquisition unit, a second distinct value acquisition unit, and an output unit;

[0129] The financial data sample acquisition unit is used to acquire candidate financial business datasets as financial data samples.

[0130] The first distinct value acquisition unit is used to determine the first distinct value of the financial data sample based on partial data of the financial data sample;

[0131] The second distinct value acquisition unit is used to determine the second distinct value of the financial data sample based on all the data of the financial data sample;

[0132] The output unit is used to determine the discriminative power of financial business data based on the comparison between the number of first distinct values ​​and the number of second distinct values.

[0133] This application also proposes a readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps:

[0134] Acquire candidate financial business datasets to create financial data samples;

[0135] Determine the number of the first distinct values ​​R1 of the financial data sample based on partial data from the financial data sample;

[0136] Determine the number of second distinct values ​​R² in the financial data sample based on all the data in the financial data sample;

[0137] The discriminant power of financial business data is determined by comparing the number of first distinct values ​​R1 and the number of second distinct values ​​R2.

[0138] This application also proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps: acquiring candidate financial business datasets and combining them into financial data samples;

[0139] Determine the number of the first distinct values ​​R1 of the financial data sample based on partial data from the financial data sample;

[0140] Determine the number of second distinct values ​​R² in the financial data sample based on all the data in the financial data sample;

[0141] The discriminant power of financial business data is determined by comparing the number of first distinct values ​​R1 and the number of second distinct values ​​R2.

[0142] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods.

[0143] Any references to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory.

[0144] Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.

[0145] Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0146] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0147] The above-described embodiments are merely examples of several implementation methods of this application. They are described in a relatively specific and detailed manner, but should not be construed as limiting the scope of this application's patent.

[0148] It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements are all within the scope of protection of this application. The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent changes made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for obtaining the distinguishability of financial business data, characterized in that, The method includes: Acquire candidate financial business datasets to create financial data samples; Determine the number of the first distinct values ​​of the financial data sample based on partial data from the financial data sample. R 1; Determine the number of second distinct values ​​for the financial data sample based on all the data in the financial data sample. R 2; Based on the number of the first distinct values R 1. Number of distinct values ​​of the second R The comparison relationship in step 2 determines the discriminative power of financial business data, specifically including: determining the data slope of the financial data sample. X ,in, X=R 2 / R 1. When the data slope X≤ When the value is fixed, use the number of the second distinct values. R 2 serves as the discriminative power of financial data samples; when the data slope X＞ When the value is fixed, the number of the second distinct values ​​is increased according to the sampling ratio. R 2. Count the number of distinct values ​​after magnification. R 2 serves as the distinguishing factor for financial data samples.

2. The method for obtaining the distinguishability of financial business data according to claim 1, characterized in that, The process of obtaining candidate financial business datasets as financial data samples specifically includes: obtaining the candidate financial business datasets as financial data samples from the database.

3. The method for obtaining the distinguishability of financial business data according to claim 2, characterized in that, The step of obtaining the candidate financial business dataset from the database as a financial data sample includes periodically updating the database or manually adding or deleting the database content.

4. The method for obtaining the distinguishability of financial business data according to claim 1, characterized in that, The step of determining the first distinct value number of the financial data sample based on partial data from the financial data sample. R 1. Specifically, this includes: selecting half of the financial data sample to determine the first distinct value number of the financial data sample. R 1.

5. The method for obtaining the distinguishability of financial business data according to claim 4, characterized in that, The method of selecting half of the financial data sample to determine the first distinct value number of the financial data sample. R 1. Specifically, it includes: from the interval [ min-id , min-id +( max-id - min-id The financial data sample is obtained from [ ) / 2], wherein, min-id The identifier number for the smallest data. max-id The identifier number for the largest data.

6. The method for obtaining the distinguishability of financial business data according to claim 1, characterized in that, The step of determining the first distinct value number of the financial data sample based on partial data from the financial data sample. R 1. Specifically, this includes: based on the screening criteria, obtaining valid data from the financial data sample as partial data to determine the first distinct value number of the financial data sample. R 1.

7. The method for obtaining the distinguishability of financial business data according to claim 1, characterized in that, The second distinct value number of the financial data sample is determined based on all the data in the financial data sample. R 2. Specifically, it includes: from the interval [ min- id , max-id All the data are obtained from [the source], where, min-id The identifier number for the smallest data. max-id The identifier number for the largest data.

8. The method for obtaining the distinguishability of financial business data according to any one of claims 1 to 7, characterized in that, The number of the first different values R 1. Number of distinct values ​​of the second R The comparison relationship in step 2, which determines the data discriminative power, also includes: automatically obtaining the number of the first distinct values ​​using functions in the database. R 1. Number of distinct values ​​of the second R 2.

9. A method or system for obtaining the distinguishability of financial business data, characterized in that, The system includes: a financial data sample acquisition unit, a first distinct value acquisition unit, a second distinct value acquisition unit, and an output unit; The financial data sample acquisition unit is used to acquire candidate financial business datasets as financial data samples; The first distinct value acquisition unit is used to determine the first distinct value of the financial data sample based on a portion of the data of the financial data sample. The second distinct value acquisition unit is used to determine the second distinct value of the financial data sample based on all the data of the financial data sample; The output unit is used to determine the discriminative power of financial business data based on the comparison relationship between the first number of different values ​​and the second number of different values. Specifically, it includes: determining the data slope X of the financial data sample, where X = R2 / R1; when the data slope X ≤ a fixed value, using the second number of different values ​​R2 as the discriminative power of the financial data sample; when the data slope X > a fixed value, amplifying the second number of different values ​​R2 according to the sampling ratio, and using the amplified second number of different values ​​R2 as the discriminative power of the financial data sample.

10. A readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the steps of the method as claimed in any one of claims 1 to 8.

11. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as claimed in any one of claims 1 to 8.

Images