Product recommendation method and device, equipment, medium and product

Abstract

The invention discloses a product recommendation method and device, equipment, a medium and a product. Relates to the technical field of computers, can be applied to the technical field of finance, and comprises the following steps: in response to a product recommendation request for a target user, determining candidate product combinations, and determining product parameter information corresponding to each candidate product combination; according to the investment constraint information of the target user, determining a target constraint condition, and according to the product parameter information corresponding to each candidate product combination, in combination with the target constraint condition, determining a revenue target function; and solving the revenue objective function based on a preset quantum approximate optimization algorithm to determine a target product combination from the candidate product combinations, and performing product recommendation on the target user. According to the technical scheme, the optimal product combination can be determined for the user by using the preset quantum approximate optimization algorithm, the recommendation efficiency is improved, meanwhile, product recommendation is better performed on the target user, and the user satisfaction is improved.

Description

Technical field

[0001] The present invention relates to the field of computer technology, is applicable to the field of financial technology, and is particularly related to a product recommendation method, device, device, medium and product. Background technology

[0002] In the current financial investment field, bank traditional product recommendation services face many challenges. On the one hand, when manually conducting market analysis and investment decisions, it is difficult to quickly and accurately process massive financial data, resulting in low efficiency in product recommendation decisions and easy to make mistakes. On the other hand, when customizing recommendation plans for customers, it is often impossible to accurately match the customer's investment constraints, which makes the recommendation results uneven.

[0003] Therefore, how to use relevant optimization algorithms to determine the optimal product combination for users, improve recommendation efficiency, and better recommend product recommendations to target users, and improve user satisfaction is a problem that needs to be solved urgently. Contents of invention

[0004] The present invention provides a product recommendation method, device, equipment, media and products to determine the optimal product combination for users, improve recommendation efficiency, and better recommend product to target users and improve user satisfaction.

[0005] According to an aspect of the invention, a product recommendation method is provided, including:

[0006] In response to a product recommendation request for the target user, a candidate product combination is determined and product parameter information corresponding to each candidate product combination is determined;

[0007] Based on the investment constraint information of the target user, the target constraint conditions are determined, and the return objective function is determined based on the product parameter information corresponding to each candidate product combination, combined with the target constraint conditions;

[0008] The profit objective function is solved based on the preset quantum approximation optimization algorithm to determine the target product combination from the candidate product combination and recommend the target user product.

[0009] According to another aspect of the invention, a product recommendation device is provided, including:

[0010] The information determination module is used to determine the product combination candidates and determine product parameter information corresponding to each product combination in response to a product recommendation request to a target user;

[0011] The function determination module is used to determine the target constraints based on the investment constraint information of the target user, and determine the return objective function based on the product parameter information corresponding to each candidate product combination, combined with the target constraints;

[0012] Recommended module is used to solve the income objective function based on the preset quantum approximation optimization algorithm to determine the target product combination from the candidate product combination and recommend the product to the target user.

[0013] According to another aspect of the invention, an electronic device is provided, which comprises:

[0014] at least one processor; and

[0015] memory in communication with the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor, the computer program executed by the at least one processor to enable the at least one processor to perform the product recommendation method of any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer readable storage medium is provided, which stores computer instructions for enabling a processor to implement the product recommendation method as described in any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, there is also provided a computer program product, which includes a computer program, and a product recommendation method for implementing any embodiment of the present invention when executed by a processor.

[0019] The technical solution of the embodiment of the present invention determines the candidate product combination in response to the product recommendation request for the target user, and determines the product parameter information corresponding to each candidate product combination; determines the target constraints based on the investment constraint information of the target user, and determines the return objective function based on the product parameter information corresponding to the candidate product combination, and combines the target constraints, and solves the return objective function based on the preset quantum approximation optimization algorithm to determine the target product combination from the candidate product combination and recommends the target user. By using preset quantum approximation optimization algorithms, the optimal product combination can be determined for users, improve recommendation efficiency and better recommend product to target users, and improve user satisfaction.

[0020] It should be understood that the description in this section is not intended to identify key or important features of embodiments of the invention, nor is it intended to limit the scope of the invention. Other features of the present invention will be easily understood by the following specification. Attached description of the drawings

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiment will be briefly introduced below. It is obvious that the accompanying drawings in the description below are only some embodiments of the invention. For those of ordinary skill in the art, other accompanying drawings can also be obtained based on these accompanying drawings without expending creative labor.

[0022] Figure 1 It is a flowchart of a product recommendation method provided by Embodiment 1 of the present invention;

[0023] Figure 2A It is a flowchart of a product recommendation method provided in Embodiment 2 of the present invention;

[0024] Figure 2B It is a schematic diagram of the calculation principle of the quantum approximation optimization algorithm provided by Embodiment 2 of the present invention;

[0025] Figure 3 It is a structural block diagram of a product recommendation device provided in Embodiment 3 of the present invention;

[0026] Figure 4 It is a structural diagram of the electronic device provided by Embodiment 4 of the present invention. Specific implementation methods

[0027] In order for those in the art to better understand the present invention solution, the technical solutions in the present invention will be clearly and completely described in light of the drawings in the present invention. It is obvious that the described embodiments are only part of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those skilled in the art without making creative labor should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", "objective", "candidate", "alternative" and the like in the specification and claims of the present invention and the appended drawings are used to distinguish similar objects, and do not necessarily use to describe a specific order or sequence. It will be understood that the data used in this way can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in order other than those illustrated or described herein. Furthermore, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, processes, methods, systems, products or devices that contain a series of steps or units are not necessarily limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. The acquisition, storage, use, and processing of data in this technical plan is in line with the relevant provisions of national laws and regulations.

[0029] It should be noted that the user information collected in the present invention 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 relevant data are subject to relevant laws, regulations and standards of relevant countries and regions, and necessary confidentiality measures are taken, and they do not violate public order and good customs. There is also a corresponding operation portal for users to choose authorization or reject automated decision-making results. If the user chooses to refuse, they will enter the expert decision-making process.

[0030] Example 1

[0031] Figure 1 It is a flowchart of a product recommendation method provided by Embodiment 1 of the present invention; This embodiment can be applied to the situation where a banking business system uses a quantum approximation optimization algorithm to recommend products to target users. The method can be performed by a product recommendation device, which can be implemented in hardware and / or software. The product recommendation device can be configured in electronic devices and executed by a bank's business system, such as Figure 1 As shown, the recommended methods for this product include:

[0032] S101. In response to a product recommendation request for the target user, a candidate product combination is determined and product parameter information corresponding to each candidate product combination is determined.

[0033] Among them, the target user may be a user who operates on the bank's business system and requests financial product recommendations, and the target user may be a customer under the banking business system. A product recommendation request refers to a request by the target user to request the banking system to recommend financial products. A candidate product portfolio refers to a product portfolio that supports selected in the business system. The candidate product portfolio can specifically be a financial investment product portfolio, such as a bond product portfolio.

[0034] The product parameter information may include at least one of the following: the average net value growth rate, product covariance, and risk weight coefficient corresponding to each candidate product portfolio. Product covariance can characterize the correlation between products in each candidate product portfolio. The average net value growth rate can characterize the rate of return corresponding to the candidate product portfolio. The risk weight coefficient can characterize the risk level of investment in candidate product portfolios, and the value range of the risk weight system can be set between 0-1.

[0035] Optionally, the target user can operate on the bank's business system to issue product recommendation instructions. When the business system detects the instruction issued by the target user, it believes that the product recommendation request for the target user has been detected. At this time, it can determine the selected product combination in the business system as a candidate product combination in response to the product recommendation request for the target user.

[0036] Optionally, the product parameter information corresponding to each candidate product combination is determined, including: vectorizing each candidate product combination, and determining the product covariance corresponding to each candidate product combination based on the vectorization results; determining the product parameter information corresponding to each candidate product combination based on the average net value growth rate, product covariance and risk weight coefficient corresponding to each candidate product combination.

[0037] Optionally, for the candidate product combination, the product characteristic information of the candidate product in each candidate product combination can be vectorized to determine the product column vector corresponding to each candidate product, and to determine the transposition of the product column vector, each row of product column vector represents a candidate product, and finally represents a vector matrix corresponding to a candidate product combination through a vector matrix. Finally, the conventional covariance calculation formula is used to determine the covariance matrix corresponding to the vector matrix to obtain the product covariance corresponding to each candidate product combination.

[0038] It should be noted that the above-mentioned solution of the present invention obtains product parameter information by considering the net value appreciation rate, covariance and risk weight of the product portfolio, and can determine more comprehensive product parameter information, so that the product portfolio can be evaluated more comprehensively, which helps to build a more reasonable objective function.

[0039] S102. Determine the target constraints based on the investment constraint information of the target user, and determine the return objective function based on the product parameter information corresponding to each candidate product combination, combined with the target constraints.

[0040] Among them, investment constraint information refers to the constraint information on the investment products that the target user can select. The investment constraint information can be the target user limits the number of selected products. For example, only 5 of the 20 products can be selected, and for example, up to five of a bond product can be selected. The target constraint is the constraint condition obtained by expressing investment constraint information through mathematical formulas. The return objective function refers to the objective function used to evaluate the expected return situation for different candidate product portfolios.

[0041] Optionally, based on the product parameter information corresponding to each candidate product combination and the target constraints, the income objective function is determined, including: based on the product parameter information corresponding to each candidate product combination and the target constraints, the initial objective function is constructed; according to the target constraints, the initial objective function is subjected to unconstrained conversion processing to obtain the unconstrained income objective function.

[0042] Exemplary, based on product parameter information corresponding to each candidate product combination, the initial objective function may be represented by the following formula:

[0043] max(μ T x-qx T ∑x)

[0044] s.t.x1+x2+…+x m =n

[0045] Among them, μ is the average net value growth rate, representing the expected rate of return of the product; ∑x is the covariance of the product, indicating the correlation between products; q is the risk weight coefficient, indicating the risk level that the investment can bear. x is the vector matrix corresponding to the candidate product combination. m is the quantity of products in the candidate product portfolio. n is the number of products in the limited product portfolio in the investment constraint information.

[0046] Exemplary, if the target user's investment constraint information is only 5 out of 20 products, the corresponding constraint condition may be x 1 +x 2 +…+x 20 =5.

[0047] It should be noted that considering that the solution of the initial objective function is a quadratic unconstrained binary optimization (QUBO), in order to facilitate the subsequent use of quantum approximation optimization algorithm for solving, the initial objective function needs to be subject to unconstrained conversion, and the constraints are added to the initial objective function to obtain the benefit objective function. Specifically, the constraints can be added to the objective function through specific mathematical transformations to eliminate the constraints of the model. For example, for the problem of finding the minimum value with the equation constraint x1+x2+x3=5, a very large constant can be specified, such as c=10000, the penalty term c(x1+x2+x3-5) can be constructed and added to the original objective function. Since the penalty term is only 0 when x1+x2+x3=5, in the optimization process, the algorithm will try its best to make the decision variables x1, x2, and x3 meet the constraints in order to obtain the minimum value during the optimization process.

[0048] Exemplary, the unconstrained benefit objective function obtained after performing unconstrained conversion of the initial objective function can be expressed as:

[0049] m ax(μ T x-qx T ∑x-c(x 1 +x 2 +…+x m -n) 2 )

[0050] Among them, μ is the average net value growth rate, representing the expected rate of return of the product; ∑x is the covariance of the product, indicating the correlation between products; q is the risk weight coefficient, indicating the risk level that the investment can bear. x is the vector matrix corresponding to the candidate product combination. m is the quantity of products in the candidate product portfolio. n is the number of products in the limited product portfolio in the investment constraint information.

[0051] It should be noted that the above-mentioned technical solution of the present invention can convert the initial objective function with the target constraints to obtain the benefit objective function. By converting the objective function into an unconstrained form, it will help to use quantum algorithms for subsequent solution, and improve the efficiency of solving the function.

[0052] Optionally, after determining the income objective function, it can be saved as a specific file, such as an LP (Linear Programming) file, for subsequent calls to the optimization solver for solving, where the LP file is a general optimization model file type with high readability.

[0053] S103. Solve the income objective function based on the preset quantum approximation optimization algorithm to determine the target product combination from the candidate product combination and recommend the target user.

[0054] Among them, the preset quantum approximation optimization algorithm is a quantum algorithm that can be used to solve problems such as combination optimization and maximum segmentation. This algorithm can map the solution of optimization problems to the problem of finding the maximum / minimum energy of the quantum system, and find the best combination from a large number of possible combinations through the energy evolution and measurement of the system. A target product portfolio refers to the product portfolio in the candidate product portfolio that maximizes the revenue objective function, that is, the product portfolio that is most suitable for the target user.

[0055] It should be noted that mathematically, the solution to the combination optimization problem can be represented by the state of n qubits, such as quantum state|0010101> In the portfolio optimization scenario, it can represent the selection of the third, fifth, and seventh investment products. Based on the quantum approximation optimization algorithm to solve the combination optimization problem, the solution of this problem must first be established in the format with the qubit state. Then, the quantum approximation optimization algorithm will continuously iterate according to the set goals to amplify the probability of the quantum state representing the optimal solution, and finally obtain the determined quantum state binary representation through measurement and decoding.

[0056] Optionally, the benefit objective function is solved based on the preset quantum approximation optimization algorithm to determine the target product combination from the candidate product combination, including: based on the preset quantum approximation optimization algorithm, convert each candidate product combination into a quantum state, and input the candidate product combination of each quantum state into a preset quantum gate array to determine the mean variance value of each candidate product combination; based on the mean variance value of each candidate product combination, and based on the preset parameter optimizer, the target quantum gate array is obtained to determine the target product combination from the candidate product combination based on the output results of the target quantum gate array.

[0057] Among them, the preset quantum approximate optimization algorithm (QAOA) is a commonly used solution algorithm. Quantum state refers to the qubit state. The candidate product combination of quantum states can be specifically expressed in binary form, such as |000> and |010>, etc. The preset quantum gate array may refer to a quantum gate array with parameters in a quantum approximation optimization algorithm. The preset parameter optimizer can be an optimizer trained based on classic parameter optimization methods. A target quantum gate array refers to a gate array that is ultimately used to determine the measurement values ​​of different candidate product combinations.

[0058] Exemplary, if the candidate product combination is a bond product combination, all possible bond combinations may be mapped to quantum states with equal probability magnitudes. Since each bond is selected at most 5 copies, the state of 3 qubits can be used to represent the selection method of 1 bond. For example: |000> means not selecting the bond, |010> means choosing 2 bonds, and so on, the combination of 5 bonds can be represented by 15 qubits.

[0059] It should be noted that the above-mentioned solution of the present invention evaluates candidate products through quantum gate arrays of quantum approximation optimization algorithms, optimizes the quantum gate arrays to determine the target quantum gate arrays for product screening, and can realize the objective function solution based on the quantum algorithm, and can quickly and effectively obtain the optimal solution of the objective function, that is, obtain the optimal target product combination.

[0060] Optionally, based on the mean analysis of variance values ​​of each candidate product combination, the target quantum gate array is obtained based on the preset parameter optimizer, including: iteratively optimize the preset parameter optimizer based on the mean analysis of variance values ​​of each candidate product combination to obtain the optimal parameters of the preset quantum gate array; substituting the optimal parameters into the preset quantum gate array to obtain the target quantum gate array.

[0061] Optionally, quantum approximation optimization algorithm lines can be loaded using the open source quantum computing programming framework, and quantum approximation optimization algorithm lines can be optimized using classic parameter optimization methods, such as Simultaneous Perturbation Stochasic Approximation, SPSA. Among them, the stochastic optimization algorithm approximates the real gradient based on stochastic gradients. Although it cannot directly obtain the exact gradient and the convergence speed is slow, it saves computing resources. In addition, SPSA only needs forward propagation to update parameters, eliminating the calculation graph required to store backpropagation, and the calculation overhead is small.

[0062] It should be noted that for quantum approximation optimization algorithm, SPSA gradually approximates the optimal solution by estimating the gradient information of the objective function. Only the estimates of two objective functions are used in each gradient approximation, regardless of the dimensions of the optimization problem. This greatly reduces the number of measurements of the objective function used to estimate gradient information, so the SPSA algorithm is suitable for solving high-dimensional problems and optimization of large-scale stochastic systems.

[0063] It should be noted that the above-mentioned technical solution of the present invention optimizes the parameter optimizer through the mean analysis of variance value to obtain the target quantum gate array based on the determined optimal parameters, so that the quantum gate array can be better optimized, so that the optimal combination can be determined from the product combination, which helps to accurately recommend the target user.

[0064] Optionally, the target product combination is determined from the candidate product combination based on the output results of the target quantum gate array, including: inputting the candidate product combination of each quantum state into the target quantum gate array to output the measured values ​​of each candidate product combination; determining the target product combination from the candidate product combination based on the measured values ​​of each candidate product combination output by the target quantum gate array.

[0065] Exemplary, a quantum gate array is constructed based on a quantum gate with parameters (quantum gate, or quantum logic gate), and the candidate product combination of each quantum state is encoded as input to the gate array. The probability amplitude of the input quantum state is changed by the quantum gate array, and the mean analysis of variance values ​​of the corresponding candidate product combination are obtained after measuring the quantum state. Then, different candidate product combinations and their mean ANOVA values ​​are sent to the classically calculated parameter optimizer to obtain the optimization value of the quantum gate array parameters, and then iterate continuously based on the new parameter values, thereby increasing the probability amplitude of the better combination. After the iteration is completed, the measurement result of the quantum circuit is the optimal product combination considered by the quantum algorithm, that is, the target product combination.

[0066] It should be noted that the above-mentioned technical solutions of the present invention determine the measurement value through the target quantum gate array to screen the product combination, and the optimal combination can be determined from the product combination, which helps to accurately recommend the target user.

[0067] The technical solution of the embodiment of the present invention determines the candidate product combination in response to the product recommendation request for the target user, and determines the product parameter information corresponding to each candidate product combination; determines the target constraints based on the investment constraint information of the target user, and determines the return objective function based on the product parameter information corresponding to the candidate product combination, and combines the target constraints, and solves the return objective function based on the preset quantum approximation optimization algorithm to determine the target product combination from the candidate product combination and recommends the target user. By using preset quantum approximation optimization algorithms, the optimal product combination can be determined for users, improve recommendation efficiency and better recommend product to target users, and improve user satisfaction.

[0068] Example 2

[0069] Figure 2A It is a flowchart of a product recommendation method provided in Embodiment 2 of the present invention; Figure 2BIt is a schematic diagram of the calculation principle of the quantum approximation optimization algorithm provided by Embodiment 2 of the present invention; Based on the above-mentioned embodiments, this embodiment provides a preferred example of product recommendation based on the quantum approximation optimization algorithm, specifically, such as Figure 2A As shown, the method includes the following process:

[0070] S201. In response to a product recommendation request for the target user, the candidate product combination is determined, and the candidate product combination is vectorized, and the product covariance corresponding to each candidate product combination is determined separately based on the vectorization results.

[0071] S202. Determine the product parameter information corresponding to each candidate product combination based on the average net value growth rate, product covariance and risk weight coefficient corresponding to each candidate product combination.

[0072] S203. Determine the target constraints based on the investment constraint information of the target user, and construct the initial objective function based on the product parameter information corresponding to each candidate product combination and combine the target constraints.

[0073] S204. According to the target constraint conditions, the initial objective function is subjected to unconstrained conversion to obtain the unconstrained return objective function.

[0074] S205. Based on the preset quantum approximation optimization algorithm, each candidate product combination is converted into a quantum state, and the candidate product combination of each quantum state is input into a preset quantum gate array to determine the mean variance analysis value of each candidate product combination.

[0075] S206. According to the mean analysis of variance values ​​of each candidate product combination, the preset parameter optimizer is iteratively optimized to obtain the optimal parameters of the preset quantum gate array.

[0076] Example, see Figure 2B , the candidate product combination of each quantum state can be input into the quantum gate array with parameters, that is, the preset quantum gate array, and according to the output results, combined with a preset parameter optimizer based on the classical parameter optimization method, the parameters of the quantum gate array are iteratively optimized to obtain new parameters, that is, the optimal parameters of the preset quantum gate array are obtained to feed back to the preset quantum gate array, thereby obtaining the target quantum gate array.

[0077] S207. Substitute the optimal parameters into the preset quantum gate array to obtain the target quantum gate array, and input the candidate product combination of each quantum state into the target quantum gate array to output the measured values ​​of each candidate product combination.

[0078] S208. The target product combination is determined from the candidate product combination based on the measured values ​​of each candidate product combination output by the target quantum gate array.

[0079] Example 3

[0080] Figure 3 It is a structural block diagram of a product recommendation device provided by Embodiment 3 of the present invention; This embodiment can be applied to the situation where a banking business system uses a quantum approximation optimization algorithm to recommend products to target users. The product recommendation device provided by Embodiment 2 of the present invention can execute the product recommendation method provided by any of the present invention, and has the corresponding functional modules and beneficial effects of the execution method; the product recommendation device can be implemented in hardware and / or software, and is configured in an electronic device with product recommendation function, and is executed by the bank's business system, such as Figure 3 As shown, the recommended device for this product specifically includes:

[0081] The information determination module 301 is configured to determine the candidate product combination in response to a product recommendation request to the target user and determine product parameter information corresponding to each candidate product combination;

[0082] The function determination module 302 is used to determine the target constraints based on the investment constraint information of the target user, and determine the return objective function based on the product parameter information corresponding to the candidate product combination, combined with the target constraints;

[0083] The recommendation module 303 is used to solve the income objective function based on the preset quantum approximation optimization algorithm to determine the target product combination from the candidate product combination and recommend the product to the target user.

[0084] The technical solution of the embodiment of the present invention determines the candidate product combination in response to the product recommendation request for the target user, and determines the product parameter information corresponding to each candidate product combination; determines the target constraints based on the investment constraint information of the target user, and determines the return objective function based on the product parameter information corresponding to the candidate product combination, and combines the target constraints, and solves the return objective function based on the preset quantum approximation optimization algorithm to determine the target product combination from the candidate product combination and recommends the target user. By using preset quantum approximation optimization algorithms, the optimal product combination can be determined for users, improve recommendation efficiency and better recommend product to target users, and improve user satisfaction.

[0085] Further, the recommendation module 303 may include:

[0086] The determination unit is used to convert each candidate product combination into a quantum state based on a preset quantum approximation optimization algorithm, and input the candidate product combination of each quantum state into a preset quantum gate array to determine the mean analysis of variance value of each candidate product combination;

[0087] The filtering unit is used to obtain a target quantum gate array based on the mean analysis of variance values ​​of each candidate product combination, based on the preset parameter optimizer, to determine the target product combination from the candidate product combination based on the output results of the target quantum gate array.

[0088] Further, the screening unit is specifically used for:

[0089] According to the mean analysis of variance values ​​of each candidate product combination, the preset parameter optimizer is iteratively optimized to obtain the optimal parameters of the preset quantum gate array;

[0090] Substitute the optimal parameters into the preset quantum gate array to obtain the target quantum gate array.

[0091] Further, the screening unit is also used for:

[0092] The candidate product combinations of each quantum state are input into the target quantum gate array to output the measured values ​​of each candidate product combination;

[0093] The target product combination is determined from the candidate product combination based on the measured values ​​of each candidate product combination output by the target quantum gate array.

[0094] Further, the information determination module 301 is specifically used for:

[0095] Each candidate product combination is vectorized, and the product covariance corresponding to each candidate product combination is determined according to the vectorization results;

[0096] The product parameter information corresponding to each candidate product combination is determined based on the average net value growth rate, product covariance and risk weight coefficient corresponding to each candidate product combination.

[0097] Further, the function determination module 302 is specifically used for:

[0098] Based on the product parameter information corresponding to each candidate product combination, combined with the target constraints, the initial objective function is constructed;

[0099] According to the target constraints, the initial objective function is subjected to unconstrained conversion to obtain the unconstrained return objective function.

[0100] Example 4

[0101] Figure 4 It is a structural diagram of the electronic device provided by Embodiment 4 of the present invention. Figure 4 A structural diagram of an electronic device 10 that may be used to implement an embodiment of the present invention is shown. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the invention described and / or claimed herein.

[0102] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by the at least one processor, and the processor 11 can perform various appropriate actions and processing according to a computer program stored in the read-only memory (ROM) 12 or a computer program loaded into the random access memory (RAM) 13 from the storage unit 18. In the RAM 13 , various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11 , the ROM 12 , and the RAM 13 are connected to each other via a bus 14 . The input / output (I / O) interface 15 is also connected to the bus 14 .

[0103] A plurality of components in the electronic device 10 are connected to the I / O interface 15 , including an input unit 16 , such as a keyboard, a mouse, etc. , an output unit 17 , such as a display, a speaker, etc. of various types, a storage unit 18 , such as a magnetic disk, an optical disk, etc. , and a communication unit 19 , such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0104] The processor 11 may be a variety of general and / or dedicated processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processors, controllers, microcontrollers, and the like. The processor 11 executes various methods and processes described above, such as product recommendation methods.

[0105] In some embodiments, the product recommendation method may be implemented as a computer program that is tangibly contained in a computer readable storage medium, such as a storage unit 18 . In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19 . When a computer program is loaded into the RAM 13 and executed by the processor 11 , one or more steps of the product recommendation method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform product recommendation methods by any other suitable means (eg, by means of firmware).

[0106] Various embodiments of the systems and techniques described above herein may be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementing in one or more computer programs that may be executed and / or interpreted on a programmable system including at least one programmable processor, the programmable processor may be a dedicated or general-purpose programmable processor that may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0107] The computer program used to implement the method of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, a dedicated computer or other programmable data processing device such that the computer program, when executed by the processor, causes the functions / operations specified in the flowchart and / or block diagrams to be implemented. The computer program may be executed entirely on the machine, partly on the machine, partly on the machine as a standalone software package and partly on the remote machine or entirely on the remote machine or server.

[0108] In the context of the present invention, a computer readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or device. Computer readable storage media may include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, semiconductor systems, devices or devices, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of machine-readable storage media would include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), fiber optic, convenient compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0109] To provide interaction with the user, the systems and techniques described herein may be implemented on an electronic device, which has a display device for displaying information to the user (eg, a CRT (cathode ray tube) or an LCD (liquid crystal display) monitor); and a keyboard and a pointing device (eg, a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (eg, visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input or, tactile input).

[0110] The systems and techniques described herein may be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., application server), or a computing system including frontend components (e.g., a user computer with a graphical user interface or web browser through which the user can interact with the implementation of the systems and techniques described herein), or in a computing system including such backend components, middleware components, or any combination of frontend components. The components of the system may be connected to each other by any form or medium digital data communication (eg, a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network and the Internet.

[0111] The computing system may include a client and a server. Clients and servers are generally far away from each other and usually interact over a communication network. The relationship between the client and the server is generated by a computer program running on the corresponding computer and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, and is a host product in the cloud computing service system to solve the shortcomings of high management difficulties and weak business expansion in traditional physical hosts and VPS services.

[0112] In one embodiment, the embodiment of the present invention further includes a computer program product, the computer program product includes a computer program, and the computer program implements the product recommendation method of any embodiment of the present invention when executed by a processor.

[0113] In the process of implementation of the computer program product, computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or combinations thereof. The programming language includes an object-oriented programming language, which can be executed entirely on the user computer, partly on the user computer, as a separate software package, partly on the user computer, partly on the remote computer, or entirely on the remote computer or server. In cases involving remote computers, the remote computer may be connected to a user computer through any kind of network—including a local area network (LAN) or wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider to connect over the Internet).

[0114] It should be understood that the various forms of flow shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention may be performed in parallel or sequentially or in different orders, and this is not limited here as long as the desired results of the technical solution of the present invention can be achieved.

[0115] The above-mentioned specific embodiments do not constitute a limitation on the scope of protection of the present invention. It will be understood by those skilled in the art that various modifications, combinations, subcombinations and substitutions can be made depending on design requirements and other factors. Any modification, equivalent substitution and improvement within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A product recommendation method, characterized in that: include: In response to a product recommendation request for a target user, determine a candidate product combination and determine product parameter information corresponding to each candidate product combination; Determine the target constraint conditions based on the target user's investment constraint information, and determine the profit objective function based on the product parameter information corresponding to each candidate product combination and the target constraint conditions; The revenue objective function is solved based on the preset quantum approximate optimization algorithm to determine the target product combination from the candidate product combination and make product recommendations to the target users.

2. The method according to claim 1, characterized in that: The revenue objective function is solved based on a preset quantum approximate optimization algorithm to determine the target product portfolio from the candidate product portfolios, including: Based on a preset quantum approximate optimization algorithm, each candidate product combination is converted into a quantum state, and the candidate product combination of each quantum state is input into a preset quantum gate array to determine the mean variance analysis value of each candidate product combination; According to the mean variance analysis value of each candidate product combination, based on a preset parameter optimizer, a target quantum gate array is obtained, so as to determine the target product combination from the candidate product combination according to the output result of the target quantum gate array.

3. The method according to claim 2, characterized in that According to the mean variance analysis values ​​of each candidate product combination, based on the preset parameter optimizer, the target quantum gate array is obtained, including: According to the mean variance analysis values ​​of each candidate product combination, the preset parameter optimizer is iteratively optimized to obtain the optimal parameters of the preset quantum gate array; Substitute the optimal parameters into the preset quantum gate array to obtain the target quantum gate array.

4. The method according to claim 2, characterized in that: According to the output results of the target quantum gate array, a target product combination is determined from the candidate product combinations, including: Inputting the candidate product combinations of each quantum state into the target quantum gate array to output the measurement value of each candidate product combination; According to the measurement values ​​of each candidate product combination output by the target quantum gate array, a target product combination is determined from the candidate product combinations.

5. The method according to claim 1, characterized in that Determine the product parameter information corresponding to each candidate product combination, including: Perform vectorization processing on each candidate product combination, and determine the product covariance corresponding to each candidate product combination according to the vectorization result; The product parameter information corresponding to each candidate product combination is determined based on the average net asset value growth rate, product covariance and risk weight coefficient corresponding to each candidate product combination.

6. The method according to claim 1, characterized in that According to the product parameter information corresponding to each candidate product combination and the target constraints, the profit objective function is determined, including: According to the product parameter information corresponding to each candidate product combination and the target constraint conditions, an initial target function is constructed; According to the objective constraints, the initial objective function is subjected to unconstrained transformation to obtain an unconstrained benefit objective function.

7. A product recommendation device, characterized in that: include: An information determination module, used to determine candidate product combinations in response to a product recommendation request for a target user, and to determine product parameter information corresponding to each candidate product combination; The function determination module is used to determine the target constraint conditions according to the investment constraint information of the target user, and determine the profit target function according to the product parameter information corresponding to each candidate product combination and the target constraint conditions; The recommendation module is used to solve the revenue objective function based on a preset quantum approximate optimization algorithm to determine the target product combination from the candidate product combination and recommend products to the target users.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the product recommendation method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the product recommendation method according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the computer program implements the product recommendation method according to any one of claims 1 to 6.

Images