Financial product determination method and device, electronic equipment and computer program product

Abstract

The invention discloses a financial product determination method and device, electronic equipment and a computer program product. Relates to the field of financial science and technology, and the method comprises the steps: obtaining a candidate financial product combination, obtaining the product data of N financial products in the candidate financial product combination, and obtaining N product data, each product data representing the income state of the financial product; a product screening model is obtained, the N pieces of product data are input into the product screening model, a product prediction label is output, and the product prediction label is obtained through prediction of Y product indexes determined by the product screening model according to the N pieces of product data; and under the condition that the product prediction tag indicates that the candidate financial product combination meets the product demand of the target user, determining the candidate financial product combination as a target financial product combination, and sending the target financial product combination to a client of the target user. According to the method and the device, the problem of low screening precision when financial products are screened in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and specifically, to a method, device, electronic device and computer program product for determining a financial product. Background Art

[0002] With the development of financial technology, support vector machines (SVMs) are now commonly used in financial analysis to construct maximum margin classifiers, effectively processing nonlinear, high-dimensional financial data. Furthermore, with the continuous advancement of machine learning research, SVMs can construct key indicators such as the historical yield and risk-adjusted return of financial products. These indicators can be used to make product predictions and provide investors with a basis for decision-making.

[0003] However, the support vector machine model still has two major shortcomings, which limit its application in investment decision analysis: first, the current support vector machine model mainly relies on the financial indicators of the fund itself when making predictions, while ignoring the decision-making ability and market adaptability of operators, which limits the model's predictive ability and makes it difficult to fully evaluate the fund's potential; in addition, financial products often exhibit an unbalanced distribution, that is, there are significant differences in the number of samples of different products. When processing unbalanced data sets, the support vector machine model tends to favor categories with a larger number of samples, resulting in a decline in performance for minority categories, which not only affects the model's generalization ability but also greatly reduces the prediction accuracy.

[0004] Regarding the technical problem of low screening accuracy when screening financial products in relevant technologies, no effective solution has been proposed so far. Summary of the Invention

[0005] The main purpose of this application is to provide a method, device, electronic device and computer program product for determining financial products to solve the technical problem of low screening accuracy when screening financial products in related technologies.

[0006] To achieve the above-mentioned objectives, according to one aspect of the present application, a method for determining a financial product is provided. The method comprises: obtaining a candidate financial product portfolio and obtaining product data for N financial products in the candidate financial product portfolio to obtain N product data, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer; obtaining a product screening model, inputting the N product data into the product screening model, and outputting a product prediction label, wherein the product prediction label is used to indicate whether the candidate financial product portfolio meets the product needs of a target user, and the product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, and Y is a positive integer; if the product prediction label indicates that the candidate financial product portfolio meets the product needs of the target user, determining the candidate financial product portfolio as a target financial product portfolio, and sending the target financial product portfolio to the client of the target user.

[0007] Furthermore, the product screening model is trained in the following manner: obtaining K historical financial product combinations, and obtaining historical product data of each historical financial product in each historical financial product combination, to obtain K groups of historical product data, wherein each historical financial product combination is associated with a product label, and each product label is used to indicate whether each historical financial product combination meets the preset requirements, and K is a positive integer; preprocessing the K groups of historical product data to obtain K groups of preprocessed product data, and calculating Y product indicators of each historical financial product combination based on each group of preprocessed product data to obtain K groups of product training indicators, wherein each group of product training indicators includes at least one of the following: logarithmic return index, adjusted return index, risk index, and and capability indicators; divide K historical financial product portfolios according to K groups of product training indicators to obtain a first product portfolio set and a second product portfolio set, determine a product training set according to the first product portfolio set and the associated product labels, and determine a product test set according to the second product portfolio set and the associated product labels; obtain a preset product screening model, and train the preset product screening model using the product training set to obtain a trained product screening model, wherein the preset product screening model is constructed according to L model parameters, and the L model parameters include at least one of the following: a preset penalty parameter and a preset bandwidth parameter, and L is a positive integer; adjust the trained product screening model using the product test set to obtain a product screening model.

[0008] Furthermore, K historical financial product combinations are divided according to K groups of product training indicators to obtain a first product combination set and a second product combination set, including: screening out the logarithmic yield indicator of each historical financial product combination from each group of product training indicators of each historical financial product combination to obtain K logarithmic yield indicators; sorting the K logarithmic yield indicators in descending order to obtain an indicator sequence, determining the historical financial product combination corresponding to the logarithmic yield indicator of a preset order in the indicator sequence as a first sample combination, and determining the historical financial product combination other than the first sample combination as a second sample combination; using a preset sampling algorithm to undersample the second sample combination to obtain a processed second sample combination; combining the first sample combination with the processed second sample combination in a preset proportion to obtain the first product combination set, and determining the historical financial product combination other than the product training set as the second product combination set.

[0009] Furthermore, Y product indicators of each historical financial product portfolio are calculated based on each group of preprocessed product data to obtain K groups of product training indicators, including: for a group of historical financial product portfolios, the unit net value of each historical financial product is obtained from a group of preprocessed product data corresponding to the historical financial product portfolio to obtain a group of unit net values, and the return value of each historical financial product is obtained to obtain a group of return values, wherein each group of unit net values ​​includes the unit net value of each historical financial product at the first moment and the unit net value of the second moment, and the first moment is the moment before the second moment; Y-1 product indicators are calculated based on a group of unit net values ​​and a group of return values, wherein the Y-1 product indicators include: a logarithmic return index, an adjusted return index and a risk index; the capability index of the product operation object of each historical financial product is obtained from a group of preprocessed product data, and the capability index of the historical financial product portfolio is determined based on the capability index of the product operation object of each historical financial product to obtain a capability combination index; the Y-1 product indicators and the capability combination index are combined to obtain a group of product training indicators of the historical financial product portfolio.

[0010] Furthermore, calculating Y-1 product indicators based on a set of unit net values ​​and a set of return values ​​includes: for a historical financial product in the historical financial product portfolio, calculating the sum of the unit net value of the historical financial product at the second moment and the return value of the historical financial product portfolio to obtain the total return; calculating the ratio of the total return to the unit net value of the historical financial product portfolio at the first moment to obtain the net value ratio, and calculating the logarithm of the net value ratio to obtain the logarithmic yield index; calculating the standard deviation of the logarithmic yield indicators of all historical financial products in the historical financial product portfolio to obtain the standard deviation of the portfolio return, and calculating the mean of the logarithmic yield indicators of all historical financial products in the historical financial product portfolio to obtain the mean portfolio return; obtaining the first preset interest rate value of the historical financial product portfolio, calculating the difference between the mean portfolio return and the preset interest rate value to obtain the return difference, calculating the ratio of the return difference to the standard deviation of the portfolio return to obtain the adjusted return index; obtaining the second preset interest rate value of the historical financial product portfolio, using the second preset interest rate value and the mean portfolio return to calculate the risk index, and the logarithmic return index, the adjusted return index and the risk index constitute Y-1 product indicators.

[0011] Further, the trained product screening model is adjusted using the product test set to obtain the product screening model, including: obtaining the recognition accuracy of the trained product screening model to obtain the control accuracy; for a historical financial product combination in the product test set, deleting any product training indicator from the product training indicators corresponding to the historical financial product combination to obtain Y-1 groups of deleted product training indicators, wherein each group of deleted product training indicators includes Y-1 product indicators; for a group of deleted product training indicators, inputting the Y-1 deleted product training indicators corresponding to each historical financial product combination in the product test set into the trained product screening model, and outputting a test label corresponding to each historical financial product combination; for a group of deleted product training indicators, obtaining a product label for each historical financial product combination in the product test set, determining the recognition accuracy based on the test label and the corresponding product label corresponding to each historical financial product combination in the product test set to obtain the test accuracy; comparing the Y-1 test accuracies with the control accuracies to obtain a comparison result, and determining the weights of the Y product indicators based on the comparison result to obtain the Y indicator weights; and adjusting the parameters of the trained product screening model based on the Y indicator weights to obtain the product screening model.

[0012] Furthermore, obtaining a candidate financial product combination includes: obtaining M candidate financial products, and obtaining product-related data for each candidate financial product, to obtain M groups of product-related data, wherein each group of product-related data includes at least one of the following: product establishment time, product launch area, and product operation object, and M is a positive integer; obtaining product screening rules, and using the product screening rules to screen each group of product-related data to obtain N groups of product-related data, wherein the product screening rules include at least one of the following: the product launch area of ​​each financial product is a preset area, the product establishment time of each financial product is within a preset time period, and the product operation object of each financial product is inconsistent; determining N candidate financial products corresponding to the N groups of product-related data, and forming the N candidate financial products into a candidate financial product combination.

[0013] To achieve the above-mentioned purpose, according to another aspect of the present application, a device for determining a financial product is provided. The device includes: a first acquisition unit, configured to acquire a candidate financial product portfolio and obtain product data of N financial products in the candidate financial product portfolio to obtain N product data, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer; a second acquisition unit, configured to acquire a product screening model, input the N product data into the product screening model, and output a product prediction label, wherein the product prediction label is used to indicate whether the candidate financial product portfolio meets the product needs of a target user, and the product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, and Y is a positive integer; and a determination unit, configured to determine the candidate financial product portfolio as a target financial product portfolio if the product prediction label indicates that the candidate financial product portfolio meets the product needs of the target user, and to send the target financial product portfolio to the client of the target user.

[0014] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is further provided, the computer-readable storage medium including a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute any of the above-mentioned methods for determining financial products.

[0015] According to another aspect of an embodiment of the present invention, an electronic device is also provided, including one or more processors and a memory, wherein the memory stores an executable program, and the processor is used to run the program, wherein when the one or more programs are executed by one or more processors, the one or more processors implement any of the above-mentioned methods for determining financial products.

[0016] According to another aspect of an embodiment of the present invention, a computer program product is provided. The computer program product includes a computer program, wherein when the computer program is executed by a processor, any one of the above-mentioned methods for determining a financial product is implemented.

[0017] In an embodiment of the present application, a method for determining financial products is adopted. By obtaining a candidate financial product portfolio and obtaining product data for N financial products in the candidate financial product portfolio, N product data are obtained, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer. A product screening model is obtained, the N product data are input into the product screening model, and a product prediction label is output, wherein the product prediction label is used to indicate whether the candidate financial product portfolio meets the product needs of a target user. The product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, and Y is a positive integer. If the product prediction label indicates that the candidate financial product portfolio meets the product needs of the target user, the candidate financial product portfolio is determined as a target financial product portfolio, and the target financial product portfolio is sent to the client of the target user. This solves the technical problem of low screening accuracy in the related art when screening financial products. By inputting the product data corresponding to the candidate financial products in the candidate financial product portfolio into the product screening model, outputting the product prediction label, and judging the candidate financial product portfolio based on the product prediction label, the technical effect of improving the accuracy of model prediction and improving the accuracy of financial product screening is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0019] Figure 1 is a hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for determining a financial product;

[0020] Figure 2 is a flowchart of a method for determining a financial product according to an embodiment of the present application;

[0021] Figure 3 is a schematic diagram of a method for determining an optional financial product provided in an embodiment of the present application;

[0022] Figure 4 is a schematic diagram of a device for determining a financial product according to an embodiment of the present application;

[0023] Figure 5 This is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set up between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain the relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0027] It should be noted that the collected information used in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse use.

[0028] Example 1

[0029] According to an embodiment of the present application, an embodiment of a method for determining a financial product is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0030] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 It is a hardware structure diagram of a computer terminal (or mobile device) for implementing a method for determining financial products, such as Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more ( Figure 1 The computer system 102 includes a processor 102 (the processor 102 may include but is not limited to a microcontroller unit (MCU) or a programmable logic device (FPGA)), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, the computer system 102 may include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS), a network interface, a keyboard, a cursor control device, a power supply, and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0031] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components of the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0032] Memory 104 can be used to store software programs and modules for application software, such as the program instructions / data storage device corresponding to the financial product determination method in the embodiments of the present application. Processor 102 executes the software programs and modules stored in memory 104 to perform various functional applications and data processing, thereby implementing the aforementioned financial product determination method. Memory 104 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 104 may further include memory remotely located from processor 102, which can be connected to computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0033] The transmission device 106 is used to receive or send data via a network. A specific example of the aforementioned network may include a wireless network provided by the communications provider of the computer terminal 10. In one embodiment, the transmission device 106 includes a network interface controller (NIC) and a network interface, which can be connected to other network devices via a base station to enable communication with the Internet. In one embodiment, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0034] The display may be, for example, a touch screen liquid crystal display (LCD), which enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0035] Under the above operating environment, this application provides Figure 2 The method for determining the financial products shown. Figure 2 is a flow chart of a method for determining a financial product according to an embodiment of the present application. Figure 2 As shown, the method includes the following steps:

[0036] Step S201: obtain a candidate financial product portfolio and obtain product data of N financial products in the candidate financial product portfolio to obtain N product data, wherein each product data represents the income status of each financial product in a preset time period, and N is a positive integer.

[0037] Specifically, in financial investment decisions, in order to screen out potential high-quality combinations from a large number of financial products, first, after obtaining a candidate financial product portfolio including multiple financial products, you can collect product data for each financial product from the database. For example, collect the income data of each financial product within a preset time period (such as one year, three years, or five years) and relevant data of the product manager who manages the product, such as years of management, historical performance, risk control capabilities, etc.

[0038] It should be noted that after obtaining the product data of these financial products, these data need to be standardized to ensure that all data are at the same level to avoid excessive influence of a certain indicator on the model. For incomplete historical data, reasonable interpolation methods or deletion strategies are used to handle missing values ​​to ensure the accuracy of model input.

[0039] Step S202: Obtain a product screening model, input N product data into the product screening model, and output a product prediction label, where the product prediction label is used to indicate whether the candidate financial product portfolio meets the product needs of the target user. The product prediction label is predicted by the product screening model based on Y product indicators determined by the N product data, where Y is a positive integer.

[0040] Specifically, the product screening model can be trained using a support vector machine model. By adding product manager competency metrics and addressing the imbalanced sample problem, the model's prediction accuracy is improved. When training the model, standardized product data is used to ensure that the model learns features relevant to the product's predicted label. Cross-validation is also employed to evaluate the model's generalization and prediction accuracy across different datasets.

[0041] After obtaining product data for the candidate financial products in the portfolio, this data is fed into a trained product screening model. Based on the input product data, the model outputs a product prediction label, providing a comprehensive assessment of whether the product meets the target user's needs. For example, a label could be classified based on whether the product meets criteria such as high return, low risk, and moderate fees, such as "investment preferred," "moderate risk," or "caution required." This prediction label allows users to quickly understand the comprehensive risk and return performance of each financial product in the portfolio, as well as whether it truly meets their investment goals and preferences.

[0042] Step S203 : when the product prediction tag indicates that the candidate financial product combination meets the product demand of the target user, the candidate financial product combination is determined as the target financial product combination, and the target financial product combination is sent to the client of the target user.

[0043] Specifically, after the product screening model generates product prediction labels based on the product data of financial products, it can match the product prediction labels with the target user's investment needs to determine which financial product combinations meet the user's specific criteria. For example, for an investor seeking high returns, if the product prediction label corresponding to a candidate financial product combination identifies the combination as "high return," it indicates that the candidate financial product combination meets the target user's product needs.

[0044] When a product prediction label clearly indicates that a candidate financial product portfolio meets the target user's product needs, that portfolio is determined to be the target financial product portfolio. However, since financial products vary with market trends, the target financial product portfolio needs to be regularly reassessed and adjusted as necessary to maintain its optimal state based on market dynamics and user needs. Furthermore, the target user can set parameters such as their investment preferences, risk tolerance, investment goals, and time horizon in the client to ensure that the recommended financial product portfolio precisely matches their needs.

[0045] The financial product determination method provided in an embodiment of the present application obtains a candidate financial product portfolio and product data of N financial products in the candidate financial product portfolio to obtain N product data, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer; obtains a product screening model, inputs the N product data into the product screening model, and outputs a product prediction label, wherein the product prediction label indicates whether the candidate financial product portfolio meets the product needs of a target user, and the product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, and Y is a positive integer; if the product prediction label indicates that the candidate financial product portfolio meets the product needs of the target user, the candidate financial product portfolio is determined as a target financial product portfolio, and the target financial product portfolio is sent to a client of the target user. This solves the technical problem of low screening accuracy in the related art when screening financial products. By inputting the product data corresponding to the candidate financial products in the candidate financial product portfolio into the product screening model, outputting the product prediction label, and judging the candidate financial product portfolio based on the product prediction label, the technical effect of improving the accuracy of model prediction and improving the accuracy of financial product screening is achieved.

[0046] Optionally, in the method for determining financial products provided in the embodiment of the present application, the product screening model is trained in the following manner: obtaining K historical financial product combinations, and obtaining historical product data of each historical financial product in each historical financial product combination, to obtain K groups of historical product data, wherein each historical financial product combination is associated with a product label, each product label is used to indicate whether each historical financial product combination meets preset requirements, and K is a positive integer; preprocessing the K groups of historical product data to obtain K groups of preprocessed product data, and calculating Y product indicators of each historical financial product combination based on each group of preprocessed product data to obtain K groups of product training indicators, wherein each group of product training indicators includes at least one of the following: logarithmic yield index, adjustment index, Return indicators, risk indicators and capability indicators; divide K historical financial product portfolios according to K groups of product training indicators to obtain a first product portfolio set and a second product portfolio set, determine a product training set according to the first product portfolio set and the associated product labels, and determine a product test set according to the second product portfolio set and the associated product labels; obtain a preset product screening model, train the preset product screening model using the product training set, and obtain a trained product screening model, wherein the preset product screening model is constructed according to L model parameters, and the L model parameters include at least one of the following: a preset penalty parameter and a preset bandwidth parameter, and L is a positive integer; adjust the trained product screening model using the product test set to obtain a product screening model.

[0047] Specifically, before using a product screening model for product prediction, the model needs to be trained. First, multiple historical financial product portfolios covering investment strategies at different time points and in different market environments, as well as different product labels, can be obtained from a historical database. Furthermore, historical product data for each historical financial product in each historical financial product portfolio can be collected to obtain the corresponding historical product data. This historical product data can include fund net value, yield, risk indicators, market environment, etc. The collected historical product data is then preprocessed to eliminate the impact of data inconsistencies on the model.

[0048] Furthermore, the preprocessed product data is used to calculate the product training indicators of each historical financial product portfolio, and based on the calculated product training indicators and the product labels associated with each historical financial product portfolio, the historical financial product portfolio is divided into a first product portfolio set and a second product portfolio set. The first product portfolio set is composed of combinations that meet preset requirements and is used to construct a product training set; the second product portfolio set is used to construct a product test set to verify the predictive ability of the model.

[0049] Furthermore, the pre-set product screening model is trained using the first product portfolio set (also known as the product training set). By adjusting model parameters, such as the penalty parameter and bandwidth parameter, the model is able to learn from historical data to effectively predict the performance of future financial product portfolios. Simultaneously, the trained product screening model is tested and adjusted using the second product portfolio set (the product test set). The model's predictive performance on new data is evaluated, and any potential overfitting or underfitting issues are identified and optimized accordingly to improve the model's generalization ability and predictive accuracy.

[0050] By training the product screening model, this embodiment can learn the advantages and disadvantages of investment strategies in different market environments, provide investors with data-based decision-making basis, and improve investment efficiency and returns.

[0051] Optionally, in the method for determining financial products provided in an embodiment of the present application, K historical financial product combinations are divided according to K groups of product training indicators to obtain a first product combination set and a second product combination set, including: screening out the logarithmic yield indicator of each historical financial product combination from each group of product training indicators of each historical financial product combination to obtain K logarithmic yield indicators; sorting the K logarithmic yield indicators in descending order to obtain an indicator sequence, determining the historical financial product combination corresponding to the logarithmic yield indicator of a preset sequence in the indicator sequence as a first sample combination, and determining the historical financial product combination other than the first sample combination as a second sample combination; using a preset sampling algorithm to undersample the second sample combination to obtain a processed second sample combination; combining the first sample combination with the processed second sample combination of a preset proportion to obtain the first product combination set, and determining the historical financial product combination other than the product training set as the second product combination set.

[0052] Specifically, to improve the model's prediction performance, the multiple historical financial product portfolios obtained can be divided. First, the logarithmic rate of return indicators directly related to investment returns are screened out from the product training indicators corresponding to each historical financial product portfolio. Then, all screened logarithmic rate of return indicators are sorted in descending order, and the logarithmic rate of return indicators with a preset rank (e.g., the top 30%) in the indicator sequence are selected. The corresponding financial product portfolios are determined as the first sample portfolio (i.e., Class 1 samples), thereby ensuring that the sample set contains the financial product portfolio with the best historical performance, thereby improving the performance of the model. The financial product portfolios other than the first sample portfolio are then determined as the second sample portfolio (i.e., Class 0 samples).

[0053] In order to solve the problem of unbalanced data in the second sample combination, a preset sampling algorithm (such as the AllKNN algorithm based on K-nearest neighbors) can be used to process the second sample combination, reduce the number of "ordinary" financial product combinations, and improve the model's ability to identify "high-quality" combinations. That is, the second sample combination is undersampled using the above algorithm to obtain a processed second sample combination. At this time, the sample combination is more balanced, which helps to improve the generalization performance of the model. Finally, the first sample combination and the processed second sample combination are combined according to a preset ratio to construct a first product combination set as the training data set of the model, and the financial product combinations other than the product training set are determined as the second product combination set. This set is used to test and verify the model to ensure the predictive ability of the model.

[0054] This embodiment constructs a more balanced and high-quality training data set by screening a high-yield historical financial product portfolio as the first sample portfolio, and combines it with the second sample portfolio after undersampling processing. This avoids overfitting of the model during training, helps to improve the model's prediction accuracy and generalization ability, and improves the model's robustness and practicality. It provides high-quality sample data for subsequent model training and testing, helping investors to more accurately evaluate the potential returns and risks of different financial product portfolios.

[0055] Optionally, in the method for determining financial products provided in an embodiment of the present application, Y product indicators of each historical financial product combination are calculated based on each group of pre-processed product data to obtain K groups of product training indicators, including: for a group of historical financial product combinations, obtaining the unit net value of each historical financial product from a group of pre-processed product data corresponding to the historical financial product combination to obtain a group of unit net values, and obtaining the return value of each historical financial product to obtain a group of return values, wherein each group of unit net values ​​includes the unit net value of each historical financial product at the first moment and the unit net value of the second moment, the first moment being the moment before the second moment; calculating Y-1 product indicators based on a group of unit net values ​​and a group of return values, wherein the Y-1 product indicators include: a logarithmic return index, an adjusted return index and a risk index; obtaining the capability index of the product operation object of each historical financial product from a group of pre-processed product data, determining the capability index of the historical financial product combination based on the capability index of the product operation object of each historical financial product, and obtaining a capability combination index; combining the Y-1 product indicators and the capability combination index to obtain a group of product training indicators for the historical financial product combination.

[0056] Before grouping multiple historical financial product portfolios, it is necessary to calculate a product training index for each portfolio and then group the different portfolios based on this index. Specifically, first, extract the unit net value of each historical financial product in each portfolio across multiple trading days. For example, the unit net value on days t and t-1 can be extracted to form a set of unit net values, where the unit net value can reflect the fluctuation in the value of the financial product. Then, obtain the return value of each historical financial product between the two aforementioned time points. For example, the return value can be the interest from day t-1 to day t.

[0057] Furthermore, the above data can be used to calculate product indicators such as logarithmic return index, adjusted return index and risk index. Although the above product indicators can analyze the investment performance and risk characteristics of financial products from different perspectives, in order to more comprehensively predict the status of financial product portfolios, the ability indicators of the product operation objects (such as product managers) of each historical financial product can be obtained, and then the ability combination indicators of the historical financial product portfolio can be determined based on the ability indicators. In this case, it can be calculated by the following formula: R p -R f =a p +β(R m -R f )+γ(R m -R f ) 2 +ε p , where a p represents the ability index, y is the timing ability index (that is, the ability of the product manager to capture market opportunities), β and γ represent the preset parameters, and R f Indicates the first preset interest rate value, R m represents the interest rate of each historical financial product, ε p Represents the preset parameters. Finally, the calculated product indicators are combined with the capability portfolio indicators to form a set of product training indicators that comprehensively reflect the historical investment performance and operational capabilities of the financial product portfolio.

[0058] This embodiment can provide data support for dividing historical financial product portfolios by calculating product training indicators, thereby laying the foundation for improving the model's prediction capabilities and enhancing the scientific nature and personalization of decision-making.

[0059] Optionally, in the method for determining financial products provided in the embodiment of the present application, calculating Y-1 product indicators based on a set of unit net values ​​and a set of yield values ​​includes: for a historical financial product in the historical financial product portfolio, calculating the sum of the unit net value of the historical financial product at the second moment and the yield value of the historical financial product portfolio to obtain the total yield; calculating the ratio of the total yield to the unit net value of the historical financial product portfolio at the first moment to obtain the net value ratio, and calculating the logarithm of the net value ratio to obtain the logarithmic yield index; calculating the standard deviation of the logarithmic yield indicators of all historical financial products in the historical financial product portfolio to obtain the standard deviation of the portfolio yield, and calculating the mean of the logarithmic yield indicators of all historical financial products in the historical financial product portfolio to obtain the mean portfolio yield; obtaining the first preset interest rate value of the historical financial product portfolio, calculating the difference between the mean portfolio yield and the preset interest rate value to obtain the yield difference, calculating the ratio of the yield difference to the standard deviation of the portfolio yield to obtain the adjusted yield index; obtaining the second preset interest rate value of the historical financial product portfolio, and using the second preset interest rate value and the mean portfolio yield to calculate the risk index, and the logarithmic yield index, the adjusted yield index and the risk index constitute Y-1 product indicators.

[0060] Specifically, when calculating product indicators such as logarithmic yield, adjusted yield, and risk, the logarithmic yield indicator can more accurately reflect the relative magnitude of changes in the value of financial products, avoiding the data skew problem that may be caused by directly using percentage yield. This indicator can be calculated using the following formula: Among them, N t and N t-1 are the unit net value of each historical financial product in each historical financial product portfolio on day t and day t-1, r t is the logarithmic return index of the historical financial product on the tth day, D t Represents the return value from day t-1 to day t. The adjusted return index is an indicator that takes both return and risk into consideration. In other words, it represents the excess return generated by a financial product for each unit of risk it bears historically. It can be calculated using the Sharpe ratio, which is the following formula: Among them, E(R p ) represents the average return of the historical financial product portfolio (i.e., the mean return of the portfolio), R f is the risk-free rate (i.e. the first preset interest rate value), σ p It represents the standard deviation of the portfolio return, which reflects the volatility of the portfolio return.

[0061] Furthermore, the adjustment of the profit index σ p The volatility of the return of the historical financial product portfolio within the T period, that is, the total risk of the historical financial product portfolio, can be calculated using the following formula: Among them, r t represents the second preset interest rate value, T represents the time period, E(R p ) represents the mean return of the portfolio.

[0062] This embodiment calculates the logarithmic rate of return index, adjusted return index and risk index, and constructs product indicators from these indicators, which can quantitatively evaluate portfolio risk, provide data support for the division of historical financial product portfolios, and also provide data support for subsequent decision-making and model optimization, thereby improving the robustness of investment strategies.

[0063] Optionally, in the method for determining financial products provided in the embodiment of the present application, the product screening model after training is adjusted using the product test set, and the product screening model is obtained, including: obtaining the recognition accuracy of the trained product screening model to obtain the control accuracy; for a historical financial product combination in the product test set, deleting any product training indicator in the product training indicators corresponding to the historical financial product combination to obtain Y-1 groups of deleted product training indicators, wherein each group of deleted product training indicators includes Y-1 product indicators; for a group of deleted product training indicators, the Y-1 deleted indicators corresponding to each historical financial product combination in the product test set are deleted. The deleted product training indicators are input into the trained product screening model, and the test labels corresponding to each historical financial product combination are output; for a set of deleted product training indicators, the product labels of each historical financial product combination in the product test set are obtained, and the recognition accuracy is determined according to the test labels and the corresponding product labels corresponding to each historical financial product combination in the product test set to obtain the test accuracy; Y-1 test accuracies are compared with the control accuracies to obtain the comparison results, and the weights of Y product indicators are determined according to the comparison results to obtain Y indicator weights; the parameters of the trained product screening model are adjusted according to the Y indicator weights to obtain the product screening model.

[0064] Specifically, when training a product screening model using a product training set, the control accuracy of the product screening model obtained using a product test set can first be obtained, where this accuracy can reflect the model's predictive performance on unseen data. Then, any indicator in the product training indicators is deleted. For example, when there are Y product indicators, deleting a certain product indicator can obtain Y-1 deleted product indicators. The remaining Y-1 product indicators are then input into the trained product screening model, and the test label output by the model is observed. This test label is then compared with the actual product label to determine the recognition accuracy of the product screening model after the missing product indicator. That is, the test accuracy is determined by using the similarity value between the test label and the product label.

[0065] Furthermore, the test accuracy of the product screening model after deleting each product indicator is compared with the control accuracy. The impact of each product indicator on model accuracy is analyzed, and an indicator weight is determined for each product indicator. If the difference between the test accuracy and the control accuracy of the model after deleting a certain indicator is large, it indicates that the indicator is crucial to model performance and can be assigned a higher weight. Conversely, if the difference between the test accuracy and the control accuracy is small, the weight of the product indicator can be appropriately reduced. Finally, these indicator weights are used to adjust the parameters of the trained product screening model, and the kernel function, penalty parameter, Gaussian bandwidth, etc. are adjusted to optimize model performance, thereby obtaining a product screening model.

[0066] This embodiment adjusts the parameters of the trained product screening model by using the indicator weights corresponding to the product indicators, which can improve the model's ability to identify and predict financial product portfolios, thereby improving the reliability of investment decisions.

[0067] Optionally, in the method for determining financial products provided in an embodiment of the present application, obtaining a candidate financial product combination includes: obtaining M candidate financial products, and obtaining product-related data for each candidate financial product to obtain M groups of product-related data, wherein each group of product-related data includes at least one of the following: product establishment time, product launch area, and product operation object, and M is a positive integer; obtaining product screening rules, and using the product screening rules to screen each group of product-related data to obtain N groups of product-related data, wherein the product screening rules include at least one of the following: the product launch area of ​​each financial product is a preset area, the product establishment time of each financial product is within a preset time period, and the product operation object of each financial product is inconsistent; determining N candidate financial products corresponding to the N groups of product-related data, and forming the N candidate financial products into a candidate financial product combination.

[0068] Specifically, when screening a portfolio of candidate financial products, you can first collect multiple candidate financial products and corresponding basic information from channels such as financial market databases and internal databases of financial institutions to obtain corresponding product-related data. For example, you can obtain product-related data such as the product establishment time, product launch area, and product operation object (that is, the product manager who operates a candidate financial product).

[0069] Furthermore, in order to reduce the impact of environmental factors, product screening rules can be used to screen financial products. That is, by setting the launch area of ​​the financial product, the establishment time of the financial product, and the product operation object, the above-mentioned multiple candidate financial products can be screened. For example, candidate financial products in region A can be selected to reduce the impact of external market fluctuations on the portfolio; financial products that have been established for more than five years can be selected to ensure that the product has a sufficiently long historical record so that its performance and stability can be effectively evaluated; financial products managed by the same operation object can be excluded to increase the diversity of the product portfolio and reduce the risks caused by decision-making errors of a single team. After eliminating financial products that do not meet the conditions in the above manner, multiple candidate financial products can be obtained, and these candidate financial products can be combined into a candidate financial product portfolio.

[0070] This embodiment uses product screening rules to select candidate financial products, which can reduce investment risks caused by market fluctuations, improve the overall quality of the investment portfolio, and make subsequent analysis and decision-making processes more focused and efficient.

[0071] This embodiment also provides a method for determining a financial product. Figure 3 is a schematic diagram of a method for determining an optional financial product provided in an embodiment of the present application, such as Figure 3 As shown, the method includes:

[0072] To screen for potential high-quality portfolios (such as funds), a product screening model constructed using a support vector machine model can be used to select appropriate financial products from a vast array of them. Before using the product screening model for product prediction, the model must be trained. First, multiple funds can be obtained and data can be selected. Specifically, these funds can be screened by setting investment regions, establishment dates, and fund managers to obtain candidate fund portfolios. Then, the fund data for each fund in the candidate portfolio is obtained and preprocessed, such as standardization, missing value processing, and dependent and independent variable processing, to obtain the corresponding product data.

[0073] Furthermore, by using the above-mentioned product data to calculate the fund performance prediction indicators of each fund, for example, the fund historical rate of return indicator, the fund risk-adjusted rate of return indicator, the fund risk indicator and the fund manager ability indicator can be calculated. A support vector machine model is constructed based on the above-mentioned indicators to obtain a trained product screening model. Finally, the prediction ability of the trained product screening model is verified. When the prediction ability meets the set threshold, the trained product screening model is put into use as the product screening model. By training the product screening model, this embodiment can learn the advantages and disadvantages of investment strategies under different market environments, provide investors with a data-based decision-making basis, and improve investment efficiency and returns.

[0074] This embodiment inputs product data corresponding to candidate financial products in a candidate financial product portfolio into a product screening model, outputs product prediction labels, and judges the candidate financial product portfolio based on the product prediction labels, thereby achieving the technical effect of improving the accuracy of model prediction and improving the accuracy of financial product screening.

[0075] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0076] Example 2

[0077] The present application also provides a device for determining a financial product. It should be noted that the device for determining a financial product in the present application can be used to execute the method for determining a financial product provided in the present application. The following describes the device for determining a financial product provided in the present application.

[0078] According to an embodiment of the present application, a device for implementing the above-mentioned method for determining a financial product is also provided. Figure 4 is a schematic diagram of a device for determining a financial product according to an embodiment of the present application, such as Figure 4 As shown, the device includes: a first acquiring unit 40, a second acquiring unit 41, and a determining unit 42.

[0079] A first acquisition unit 40 is configured to acquire a candidate financial product portfolio and obtain product data of N financial products in the candidate financial product portfolio to obtain N product data, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer;

[0080] A second acquisition unit 41 is configured to acquire a product screening model, input N product data into the product screening model, and output a product prediction label, wherein the product prediction label indicates whether the candidate financial product portfolio meets the product needs of the target user. The product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, where Y is a positive integer.

[0081] The determining unit 42 is configured to determine the candidate financial product combination as a target financial product combination when the product prediction tag indicates that the candidate financial product combination meets the product demand of the target user, and send the target financial product combination to the client of the target user.

[0082] The financial product determination device provided in the embodiment of the present application uses a first acquisition unit 40 to acquire a candidate financial product combination and obtain product data of N financial products in the candidate financial product combination to obtain N product data, where each product data represents the return status of each financial product in a preset time period, and N is a positive integer. A second acquisition unit 41 acquires a product screening model, inputs the N product data into the product screening model, and outputs a product prediction label, where the product prediction label indicates whether the candidate financial product combination meets the product needs of a target user. The product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, where Y is a positive integer. If the product prediction label indicates that the candidate financial product combination meets the product needs of the target user, the determination unit 42 determines the candidate financial product combination as a target financial product combination and sends the target financial product combination to the client of the target user. This solves the technical problem of low screening accuracy in financial product screening in related technologies. By inputting the product data corresponding to the candidate financial products in the candidate financial product combination into the product screening model, outputting the product prediction label, and judging the candidate financial product combination based on the product prediction label, the technical effect of improving the accuracy of model prediction and improving the accuracy of financial product screening is achieved.

[0083] Optionally, in the financial product determination device provided in the embodiment of the present application, the second acquisition unit 41 includes: a first acquisition module, used to obtain K historical financial product combinations, and obtain historical product data of each historical financial product in each historical financial product combination to obtain K groups of historical product data, wherein each historical financial product combination is associated with a product label, each product label is used to indicate whether each historical financial product combination meets the preset requirements, and K is a positive integer; a first processing module, used to preprocess the K groups of historical product data to obtain K groups of preprocessed product data, and calculate Y product indicators of each historical financial product combination based on each group of preprocessed product data to obtain K groups of product training indicators, wherein each group of product training indicators includes at least one of the following: logarithmic yield indicator, adjusted yield indicator, risk risk indicators and capability indicators; a partitioning module, used to partition K historical financial product portfolios according to K groups of product training indicators to obtain a first product portfolio set and a second product portfolio set, determine a product training set according to the first product portfolio set and associated product labels, and determine a product test set according to the second product portfolio set and associated product labels; a second acquisition module, used to obtain a preset product screening model, train the preset product screening model using the product training set, and obtain a trained product screening model, wherein the preset product screening model is constructed according to L model parameters, and the L model parameters include at least one of the following: a preset penalty parameter and a preset bandwidth parameter, and L is a positive integer; an adjustment module, used to adjust the trained product screening model using the product test set to obtain a product screening model.

[0084] Optionally, in the financial product determination device provided in the embodiment of the present application, the second acquisition unit 41 includes: a screening module, used to screen out the logarithmic yield index of each historical financial product combination from each group of product training indicators of each historical financial product combination, and obtain K logarithmic yield indicators; a sorting module, used to sort the K logarithmic yield indicators in descending order to obtain an indicator sequence, and determine the historical financial product combination corresponding to the logarithmic yield index of a preset sequence in the indicator sequence as a first sample combination, and determine the historical financial product combination other than the first sample combination as a second sample combination; a second processing module, used to use a preset sampling algorithm to undersample the second sample combination to obtain a processed second sample combination; a first combination module, used to combine the first sample combination of a preset proportion and the processed second sample combination of a preset proportion to obtain a first product combination set, and determine the historical financial product combination other than the product training set as a second product combination set.

[0085] Optionally, in the financial product determination device provided in the embodiment of the present application, the second acquisition unit 41 includes: a third acquisition module, which is used to obtain the unit net value of each historical financial product from a set of pre-processed product data corresponding to the historical financial product combination for a group of historical financial product combinations, to obtain a set of unit net values, and obtain the return value of each historical financial product, to obtain a set of return values, wherein each set of unit net values ​​includes the unit net value of each historical financial product at the first moment and the unit net value at the second moment, and the first moment is the moment before the second moment; a calculation module, which is used to calculate Y-1 product indicators based on a set of unit net values ​​and a set of return values, wherein the Y-1 product indicators include: a logarithmic yield index, an adjusted return index and a risk index; a fourth acquisition module, which is used to obtain the capability index of the product operation object of each historical financial product from a set of pre-processed product data, determine the capability index of the historical financial product combination based on the capability index of the product operation object of each historical financial product, and obtain a capability combination index; a second combination module, which is used to combine the Y-1 product indicators and the capability combination index to obtain a set of product training indicators for the historical financial product combination.

[0086] Optionally, in the financial product determination device provided in the embodiment of the present application, the second acquisition unit 41 includes: for a historical financial product in the historical financial product portfolio, calculating the sum of the unit net value of the historical financial product at the second moment and the return value of the historical financial product portfolio to obtain the total return; calculating the ratio of the total return to the unit net value of the historical financial product portfolio at the first moment to obtain the net value ratio, and calculating the logarithm of the net value ratio to obtain the logarithmic yield index; calculating the standard deviation of the logarithmic yield indicators of all historical financial products in the historical financial product portfolio to obtain the standard deviation of the portfolio return, and calculating the mean of the logarithmic yield indicators of all historical financial products in the historical financial product portfolio to obtain the mean portfolio return; obtaining the first preset interest rate value of the historical financial product portfolio, calculating the difference between the mean portfolio return and the preset interest rate value to obtain the return difference, calculating the ratio of the return difference to the standard deviation of the portfolio return to obtain the adjusted return index; obtaining the second preset interest rate value of the historical financial product portfolio, using the second preset interest rate value and the portfolio return mean to calculate the risk index, and the logarithmic return index, the adjusted return index and the risk index constitute Y-1 product indicators.

[0087] Optionally, in the financial product determination device provided in the embodiment of the present application, the second acquisition unit 41 includes: acquiring the recognition accuracy of the trained product screening model to obtain a control accuracy; for a historical financial product combination in the product test set, deleting any product training indicator in the product training indicators corresponding to the historical financial product combination to obtain Y-1 groups of deleted product training indicators, wherein each group of deleted product training indicators includes Y-1 product indicators; for a group of deleted product training indicators, inputting the Y-1 deleted product training indicators corresponding to each historical financial product combination in the product test set into the trained product screening model, and outputting a test label corresponding to each historical financial product combination; for a group of deleted product training indicators, acquiring a product label for each historical financial product combination in the product test set, determining the recognition accuracy based on the test label and the corresponding product label corresponding to each historical financial product combination in the product test set to obtain a test accuracy; comparing the Y-1 test accuracies with the control accuracies to obtain a comparison result, and determining the weights of the Y product indicators based on the comparison result to obtain Y indicator weights; and adjusting the parameters of the trained product screening model based on the Y indicator weights to obtain a product screening model.

[0088] Optionally, in the financial product determination device provided in the embodiment of the present application, the first acquisition unit 40 includes: acquiring M candidate financial products, and acquiring product-related data of each candidate financial product to obtain M groups of product-related data, wherein each group of product-related data includes at least one of the following: product establishment time, product launch area, and product operation object, and M is a positive integer; acquiring product screening rules, and using the product screening rules to screen each group of product-related data to obtain N groups of product-related data, wherein the product screening rules include at least one of the following: the product launch area of ​​each financial product is a preset area, the product establishment time of each financial product is within a preset time period, and the product operation object of each financial product is inconsistent; determining N candidate financial products corresponding to the N groups of product-related data, and forming the N candidate financial products into a candidate financial product combination.

[0089] It should be noted that the first acquisition unit 40, the second acquisition unit 41, and the determination unit 42 described above correspond to steps S201 to S203 in Example 1. The examples and application scenarios implemented by the above units and the corresponding steps are the same, but are not limited to the contents disclosed in the above Example 1. It should be noted that the above modules or units can be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above units can also be part of a device and can be run in the computer terminal 10 provided in Example 1.

[0090] Example 3

[0091] The embodiment of the present application may provide a computer terminal, which may be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal may also be replaced by a terminal device such as a mobile terminal or an electronic device.

[0092] Optionally, in this embodiment, the computer terminal may be located in at least one network device among a plurality of network devices of a computer network.

[0093] In this embodiment, the computer terminal may execute the program code of the following steps in the method for determining a financial product: obtaining a candidate financial product combination and obtaining product data of N financial products in the candidate financial product combination to obtain N product data, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer; obtaining a product screening model, inputting the N product data into the product screening model, and outputting a product prediction label, wherein the product prediction label is used to indicate whether the candidate financial product combination meets the product needs of the target user, and the product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, and Y is a positive integer; if the product prediction label indicates that the candidate financial product combination meets the product needs of the target user, the candidate financial product combination is determined as the target financial product combination, and the target financial product combination is sent to the client of the target user.

[0094] Optionally, the computer terminal may execute the program code of the following steps in the method for determining a financial product: obtaining K historical financial product combinations, and obtaining historical product data of each historical financial product in each historical financial product combination to obtain K groups of historical product data, wherein each historical financial product combination is associated with a product label, and each product label is used to indicate whether each historical financial product combination meets preset requirements, and K is a positive integer; preprocessing the K groups of historical product data to obtain K groups of preprocessed product data, and calculating Y product indicators of each historical financial product combination based on each group of preprocessed product data to obtain K groups of product training indicators, wherein each group of product training indicators includes at least one of the following: a logarithmic yield indicator, an adjusted yield indicator , risk indicators and capability indicators; divide K historical financial product portfolios according to K groups of product training indicators to obtain a first product portfolio set and a second product portfolio set, determine a product training set according to the first product portfolio set and the associated product labels, and determine a product test set according to the second product portfolio set and the associated product labels; obtain a preset product screening model, train the preset product screening model with the product training set, and obtain a trained product screening model, wherein the preset product screening model is constructed according to L model parameters, and the L model parameters include at least one of the following: a preset penalty parameter and a preset bandwidth parameter, and L is a positive integer; adjust the trained product screening model with the product test set to obtain a product screening model.

[0095] Optionally, the computer terminal may execute the program code of the following steps in the method for determining financial products: screening out the logarithmic yield index of each historical financial product combination from each group of product training indicators of each historical financial product combination to obtain K logarithmic yield indicators; sorting the K logarithmic yield indicators in descending order to obtain an indicator sequence, determining the historical financial product combination corresponding to the logarithmic yield index of a preset sequence in the indicator sequence as a first sample combination, and determining the historical financial product combination other than the first sample combination as a second sample combination; performing under-sampling processing on the second sample combination using a preset sampling algorithm to obtain a processed second sample combination; combining the first sample combination with the processed second sample combination in a preset proportion to obtain a first product combination set, and determining the historical financial product combination other than the product training set as a second product combination set.

[0096] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the method for determining financial products: for a group of historical financial product portfolios, obtain the unit net value of each historical financial product from a group of pre-processed product data corresponding to the historical financial product portfolio to obtain a group of unit net values, and obtain the return value of each historical financial product to obtain a group of return values, wherein each group of unit net values ​​includes the unit net value of each historical financial product at the first moment and the unit net value at the second moment, and the first moment is the moment before the second moment; calculate Y-1 product indicators based on a group of unit net values ​​and a group of return values, wherein the Y-1 product indicators include: a logarithmic return index, an adjusted return index and a risk index; obtain the capability index of the product operation object of each historical financial product from a group of pre-processed product data, determine the capability index of the historical financial product portfolio based on the capability index of the product operation object of each historical financial product, and obtain a capability combination index; combine the Y-1 product indicators and the capability combination index to obtain a group of product training indicators for the historical financial product portfolio.

[0097] Optionally, the computer terminal can execute the program code of the following steps in the method for determining financial products: for a historical financial product in a historical financial product portfolio, calculate the sum of the unit net value of the historical financial product at the second moment and the return value of the historical financial product portfolio to obtain the total return; calculate the ratio of the total return to the unit net value of the historical financial product portfolio at the first moment to obtain the net value ratio, and calculate the logarithm of the net value ratio to obtain the logarithmic yield index; calculate the standard deviation of the logarithmic yield index of all historical financial products in the historical financial product portfolio to obtain the standard deviation of the portfolio return, and calculate the mean of the logarithmic yield index of all historical financial products in the historical financial product portfolio to obtain the mean portfolio return; obtain the first preset interest rate value of the historical financial product portfolio, calculate the difference between the mean portfolio return and the preset interest rate value to obtain the return difference, calculate the ratio of the return difference to the standard deviation of the portfolio return to obtain the adjusted return index; obtain the second preset interest rate value of the historical financial product portfolio, and use the second preset interest rate value and the mean portfolio return to calculate the risk index, and the logarithmic return index, the adjusted return index and the risk index constitute Y-1 product indicators.

[0098] Optionally, the computer terminal may execute the program code of the following steps in the method for determining financial products: obtaining the recognition accuracy of the trained product screening model to obtain a control accuracy; for a historical financial product combination in a product test set, deleting any product training indicator from the product training indicators corresponding to the historical financial product combination to obtain Y-1 groups of deleted product training indicators, wherein each group of deleted product training indicators includes Y-1 product indicators; for a group of deleted product training indicators, inputting the Y-1 deleted product training indicators corresponding to each historical financial product combination in the product test set into the trained product screening model, and outputting a test label corresponding to each historical financial product combination; for a group of deleted product training indicators, obtaining a product label for each historical financial product combination in the product test set, determining the recognition accuracy based on the test label and the corresponding product label corresponding to each historical financial product combination in the product test set to obtain a test accuracy; comparing the Y-1 test accuracies with the control accuracies to obtain a comparison result, and determining the weights of the Y product indicators based on the comparison result to obtain Y indicator weights; and adjusting the parameters of the trained product screening model based on the Y indicator weights to obtain a product screening model.

[0099] Optionally, the computer terminal may execute the program code of the following steps in the method for determining a financial product: obtaining M candidate financial products, and obtaining product-related data for each candidate financial product to obtain M groups of product-related data, wherein each group of product-related data includes at least one of the following: product establishment time, product launch area, and product operation object, and M is a positive integer; obtaining product screening rules, and using the product screening rules to screen each group of product-related data to obtain N groups of product-related data, wherein the product screening rules include at least one of the following: the product launch area of ​​each financial product is a preset area, the product establishment time of each financial product is within a preset time period, and the product operation object of each financial product is inconsistent; determining N candidate financial products corresponding to the N groups of product-related data, and forming the N candidate financial products into a candidate financial product portfolio.

[0100] Optionally, Figure 5 This is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 5 As shown, the electronic device may include: one or more ( Figure 5 Only one is shown) processor 502, memory 504, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0101] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the financial product determination method and apparatus in the embodiments of the present application. The processor executes the software programs and modules stored in the memory to perform various functional applications and data processing, thereby implementing the aforementioned financial product determination method. The memory can include high-speed random access memory (RAM) and can also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory can further include memory remotely located relative to the processor, and such remote memory can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0102] The processor may call the information and application programs stored in the memory through the transmission device to execute the above steps in the above method for determining financial products.

[0103] An embodiment of the present application provides a solution for determining financial products. By obtaining a candidate financial product portfolio and obtaining product data for N financial products in the candidate financial product portfolio, N product data are obtained, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer; a product screening model is obtained, the N product data are input into the product screening model, and a product prediction label is output, wherein the product prediction label is used to indicate whether the candidate financial product portfolio meets the product needs of a target user, and the product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, and Y is a positive integer; if the product prediction label indicates that the candidate financial product portfolio meets the product needs of the target user, the candidate financial product portfolio is determined as a target financial product portfolio, and the target financial product portfolio is sent to the client of the target user, thereby achieving the technical effect of improving the accuracy of model prediction and improving the accuracy of financial product screening, thereby solving the technical problem of low screening accuracy when screening financial products.

[0104] It can be understood by those skilled in the art that Figure 5 The structure shown is for illustration only, and the electronic device may also be a terminal device such as a smart phone, a tablet computer, a PDA, a mobile Internet device (MID), or a PAD. Figure 5 It does not limit the structure of the above electronic device. For example, the electronic device may also include Figure 5 More or fewer components (such as network interfaces, display devices, etc.) shown in, or with Figure 5 Different configurations shown.

[0105] A person skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0106] Example 4

[0107] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the method for determining a financial product provided in the first embodiment.

[0108] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0109] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: obtaining a candidate financial product combination, and obtaining product data of N financial products in the candidate financial product combination to obtain N product data, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer; obtaining a product screening model, inputting the N product data into the product screening model, and outputting a product prediction label, wherein the product prediction label is used to indicate whether the candidate financial product combination meets the product needs of the target user, and the product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, and Y is a positive integer; when the product prediction label indicates that the candidate financial product combination meets the product needs of the target user, the candidate financial product combination is determined as the target financial product combination, and the target financial product combination is sent to the client of the target user.

[0110] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of the method for determining a financial product.

[0111] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0112] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0113] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0114] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0115] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0116] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program code.

[0117] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining a financial product, characterized in that: include: Obtain a candidate financial product portfolio, and obtain product data of N financial products in the candidate financial product portfolio to obtain N product data, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer; Obtaining a product screening model, inputting the N product data into the product screening model, and outputting a product prediction label, wherein the product prediction label is used to indicate whether the candidate financial product combination meets the product needs of the target user, and the product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, where Y is a positive integer; If the product prediction tag indicates that the candidate financial product combination meets the product demand of the target user, the candidate financial product combination is determined as a target financial product combination, and the target financial product combination is sent to the client of the target user.

2. The method according to claim 1, characterized in that The product screening model is trained in the following way: Obtain K historical financial product combinations, and obtain historical product data for each historical financial product in each historical financial product combination, to obtain K groups of historical product data, wherein each historical financial product combination is associated with a product tag, each product tag is used to indicate whether each historical financial product combination meets preset requirements, and K is a positive integer; Preprocessing the K groups of historical product data to obtain K groups of preprocessed product data, and calculating the Y product indicators of each historical financial product portfolio based on each group of preprocessed product data to obtain K groups of product training indicators, wherein each group of product training indicators includes at least one of the following: a logarithmic rate of return indicator, an adjusted return indicator, a risk indicator, and a capability indicator; Dividing the K historical financial product combinations according to the K groups of product training indicators to obtain a first product combination set and a second product combination set, determining a product training set based on the first product combination set and associated product labels, and determining a product test set based on the second product combination set and associated product labels; Obtaining a preset product screening model, and training the preset product screening model using the product training set to obtain a trained product screening model, wherein the preset product screening model is constructed according to L model parameters, and the L model parameters include at least one of the following: a preset penalty parameter and a preset bandwidth parameter, where L is a positive integer; The trained product screening model is adjusted using the product test set to obtain the product screening model.

3. The method according to claim 2, characterized in that The K historical financial product combinations are divided according to the K groups of product training indicators to obtain a first product combination set and a second product combination set including: Filter out the logarithmic yield index of each historical financial product portfolio from each set of product training indicators of each historical financial product portfolio to obtain K logarithmic yield indicators; Sorting the K logarithmic yield indicators in descending order to obtain an indicator sequence, determining the historical financial product combination corresponding to the logarithmic yield indicator at a preset order in the indicator sequence as a first sample combination, and determining the historical financial product combination other than the first sample combination as a second sample combination; Performing undersampling processing on the second sample combination using a preset sampling algorithm to obtain a processed second sample combination; The first sample combination of a preset proportion and the processed second sample combination of the preset proportion are combined to obtain the first product combination set, and the historical financial product combination other than the product training set is determined as the second product combination set.

4. The method according to claim 2, characterized in that The Y product indicators of each historical financial product portfolio are calculated based on each set of pre-processed product data to obtain K sets of product training indicators including: For a group of historical financial product combinations, obtaining the unit net value of each historical financial product from a set of pre-processed product data corresponding to the historical financial product combinations to obtain a set of unit net values, and obtaining the return value of each historical financial product to obtain a set of return values, wherein each set of unit net values ​​includes the unit net value of each historical financial product at a first moment and the unit net value at a second moment, where the first moment is a moment before the second moment; Calculating Y-1 product indicators based on the set of unit net values ​​and the set of return values, wherein the Y-1 product indicators include: the logarithmic return index, the adjusted return index, and the risk index; Obtaining a capability indicator of a product operation object of each historical financial product from the set of pre-processed product data, determining a capability indicator of the historical financial product combination based on the capability indicator of the product operation object of each historical financial product, and obtaining a capability combination indicator; The Y-1 product indicators and the capability combination indicator are combined to obtain a set of product training indicators for the historical financial product combination.

5. The method according to claim 4, characterized in that Calculating Y-1 product indicators based on the set of unit net values ​​and the set of return values ​​includes: For a historical financial product in the historical financial product portfolio, calculating the sum of the unit net value of the historical financial product at the second moment and the income value of the historical financial product portfolio to obtain a total income; Calculating the ratio of the total income to the unit net value of the historical financial product portfolio at the first moment to obtain a net value ratio, and calculating the logarithmic value of the net value ratio to obtain the logarithmic rate of return indicator; Calculating the standard deviation of the logarithmic rate of return indicators of all historical financial products in the historical financial product portfolio to obtain the standard deviation of the portfolio return, and calculating the mean of the logarithmic rate of return indicators of all historical financial products in the historical financial product portfolio to obtain the mean portfolio return; Obtaining a first preset interest rate value of the historical financial product portfolio, calculating a difference between the portfolio return mean and the preset interest rate value to obtain a return difference, and calculating a ratio of the return difference to the portfolio return standard deviation to obtain the adjusted return indicator; Obtain a second preset interest rate value of the historical financial product portfolio, use the second preset interest rate value and the average of the portfolio returns to calculate the risk index, and the logarithmic return index, the adjusted return index and the risk index constitute the Y-1 product indicators.

6. The method according to claim 2, characterized in that The trained product screening model is adjusted using the product test set to obtain the product screening model, including: Obtaining the recognition accuracy of the trained product screening model to obtain a control accuracy; For a historical financial product combination in the product test set, delete any product training indicator from the product training indicators corresponding to the historical financial product combination to obtain Y-1 groups of deleted product training indicators, wherein each group of deleted product training indicators includes Y-1 product indicators; For a set of deleted product training indicators, input Y-1 deleted product training indicators corresponding to each historical financial product combination in the product test set into the trained product screening model, and output a test label corresponding to each historical financial product combination; For a set of deleted product training indicators, obtain a product label for each historical financial product combination in the product test set, and determine a recognition accuracy based on the test label and the corresponding product label corresponding to each historical financial product combination in the product test set to obtain a test accuracy; Comparing Y-1 test accuracies with the control accuracies to obtain a comparison result, and determining weights of the Y product indicators based on the comparison result to obtain Y indicator weights; The parameters of the trained product screening model are adjusted according to the Y indicator weights to obtain the product screening model.

7. The method according to claim 1, characterized in that Obtain candidate financial product portfolio including: Obtain M candidate financial products and obtain product-related data for each candidate financial product, thereby obtaining M sets of product-related data, where each set of product-related data includes at least one of the following: product establishment time, product launch region, and product operation target, where M is a positive integer; Obtain product screening rules, and use the product screening rules to screen each set of product-related data to obtain N sets of product-related data, wherein the product screening rules include at least one of the following: the product launch area of ​​each financial product is a preset area, the product establishment time of each financial product is within a preset time period, and the product operation objects of each financial product are inconsistent; N candidate financial products corresponding to the N groups of product association data are determined, and the N candidate financial products are used to form the candidate financial product combination.

8. A device for determining a financial product, characterized in that: include: A first acquisition unit is configured to acquire a candidate financial product portfolio and obtain product data of N financial products in the candidate financial product portfolio to obtain N product data, wherein each product data represents the return status of each financial product in a preset time period, and N is a positive integer; a second acquisition unit, configured to acquire a product screening model, input the N product data into the product screening model, and output a product prediction label, wherein the product prediction label indicates whether the candidate financial product portfolio meets the product needs of the target user, and the product prediction label is predicted by Y product indicators determined by the product screening model based on the N product data, where Y is a positive integer; A determining unit is configured to determine the candidate financial product combination as a target financial product combination if the product prediction tag indicates that the candidate financial product combination meets the product demand of the target user, and send the target financial product combination to the client of the target user.

9. An electronic device, characterized in that: include: a memory storing an executable program; A processor is configured to run the program, wherein the program, when running, executes the method for determining a financial product according to any one of claims 1 to 7.

10. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the steps of the method for determining a financial product according to any one of claims 1 to 7 are implemented.

Images