Resource allocation method and device based on distributed constraint optimization and electronic equipment

Abstract

The invention discloses a resource allocation method and device based on distributed constraint optimization and electronic equipment, and relates to the technical field of data processing. The method comprises the following steps: matching historical service data of a target user with a knowledge graph of resource allocation to obtain at least two basic entities, and constructing and displaying an initial network based on key entities to which the basic entities belong and edge relationships among different basic entities; obtaining a target network in response to an editing operation on the initial network; constructing a distributed constraint optimization model according to the target network and the historical service data; and solving the distributed constraint optimization model to obtain a resource allocation scheme of the target user. According to the technical scheme, the accuracy and efficiency of resource allocation are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer technology, especially the field of data processing technology. Specifically, it relates to a resource allocation method, device and electronic device based on distributed constraint optimization. Background Art

[0002] In complex business scenarios, resource allocation often faces the problem of coordinated scheduling of multiple types of resources. There are complex and diverse constraint relationships between different resource types, such as the ratio limit of computing power and storage, and the dynamic coupling of network bandwidth and power supply. Moreover, resource requirements change with business fluctuations, showing time-varying characteristics. The interweaving of such multi-dimensional constraints and dynamic demands makes it difficult for traditional centralized allocation strategies to adapt. They often sacrifice overall efficiency due to local optimization, and the problems of resource fragmentation and supply-demand mismatch become more prominent.

[0003] Taking the wealth management scenario as an example, user resources need to be allocated to multiple resource types such as stocks, bonds, wealth management products, and gold derivatives. The risk-return characteristics, liquidity constraints, and market correlations of different resources form a complex configuration constraint network. How to achieve real-time response and global optimization of resource allocation through dynamic modeling has become a key challenge for improving management efficiency. Summary of the Invention

[0004] The present application provides a resource allocation method, device and electronic device based on distributed constraint optimization to improve the accuracy of abnormal problem positioning.

[0005] According to one aspect of the present application, a resource allocation method based on distributed constraint optimization is provided, including:

[0006] Matching the historical business data of the target user with the knowledge graph of resource allocation to obtain at least two basic entities, and constructing and displaying an initial network based on the key entities to which the basic entities belong and the edge relationships between different basic entities;

[0007] Responding to an editing operation on the initial network to obtain a target network;

[0008] Constructing a distributed constraint optimization model according to the target network and the historical business data;

[0009] Solving the distributed constraint optimization model to obtain a resource allocation plan for the target user.

[0010] According to another aspect of the present application, a resource allocation device based on distributed constraint optimization is provided. The device includes:

[0011] An initial network module, configured to match the historical service data of a target user with a knowledge graph of resource allocation to obtain at least two basic entities, and construct and display an initial network based on the key entities to which the basic entities belong and the edge relationships between different basic entities;

[0012] A target network module, configured to obtain a target network in response to an editing operation on the initial network;

[0013] A distributed constraint optimization module, configured to construct a distributed constraint optimization model according to the target network and the historical service data;

[0014] A resource allocation scheme module, configured to solve the distributed constraint optimization model to obtain a resource allocation scheme for the target user.

[0015] According to another aspect of the present application, there is provided an electronic device, which includes:

[0016] One or more processors;

[0017] A memory, configured to store one or more programs;

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement any method provided by the embodiments of the present application.

[0019] According to another aspect of the present application, there is provided a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements any method provided by the embodiments of the present application.

[0020] According to another aspect of the present application, there is provided a computer program product, including a computer program, and when the computer program is executed by a processor, it implements any method provided by the embodiments of the present application.

[0021] By matching the historical service data of the target user with the basic entities in the knowledge graph of resource allocation, and constructing an initial network according to the successfully matched basic entities and the key entities to which the basic entities belong, the present application can present the logic of resource constraints; and, through the visual network editing function, it supports flexible adjustment of the initial network to obtain a target network, further improving the accuracy and flexibility of the target network; constructing and solving a distributed constraint optimization model for the target network and the historical service data of the target user to obtain a resource allocation scheme for the target user can take into account the integrity of the global optimal solution and the constraint changes brought by business fluctuations, making the resource allocation scheme have both theoretical optimality and practical executability, and significantly improving the decision-making efficiency and business response speed. Description of the Drawings

[0022] Figure 1aIt is a flowchart of a resource allocation method based on distributed constraint optimization provided in Embodiment 1 of the present application;

[0023] Figure 1b It is a schematic structural diagram of an initial network provided in Embodiment 1 of the present application;

[0024] Figure 2 It is a flowchart of a resource allocation method based on distributed constraint optimization provided in Embodiment 2 of the present application;

[0025] Figure 3 It is a flowchart of a resource allocation method based on distributed constraint optimization provided in Embodiment 2 of the present application;

[0026] Figure 4 It is a schematic structural diagram of a resource allocation device based on distributed constraint optimization provided in Embodiment 3 of the present application;

[0027] Figure 5 It is a schematic structural diagram of an electronic device for implementing the resource allocation method based on distributed constraint optimization of the embodiments of the present application. Detailed implementation manners

[0028] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0029] 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 do not necessarily need to describe a specific order or sequence. It should be understood that such data used can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily need to be limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0030] In addition, it should be further noted that in the technical solution of the present application, the collection, storage, use, processing, transmission, provision and disclosure of relevant data such as the email address blacklist and the emails to be detected comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0031] Example 1

[0032] Figure 1a FIG. is a flowchart of a resource allocation method based on distributed constraint optimization according to Example 1 of the present application. This example is applicable to the situation of allocating user resources and can be executed by a resource allocation device based on distributed constraint optimization. The resource allocation device based on distributed constraint optimization can be implemented in the form of hardware and / or software, and the device can be configured in a computer device. As Figure 1a shown, the method includes:

[0033] S110. Match the historical service data of the target user with the knowledge graph of resource allocation to obtain at least two basic entities, and construct and display an initial network based on the key entities to which the basic entities belong and the edge relationships between different basic entities;

[0034] S120. Obtain a target network in response to an editing operation on the initial network;

[0035] S130. Construct a distributed constraint optimization model according to the target network and the historical service data;

[0036] S140. Solve the distributed constraint optimization model to obtain a resource allocation scheme for the target user.

[0037] During the process of allocating resources for a certain user, the user is the target user. The historical service data of the target user can be obtained from the service system by using the identifier of the target user. Among them, the historical service data can include historical transaction records, risk preference questionnaires, health monitoring data, family structure data, etc. The knowledge graph of resource allocation can include different basic entities, and the attributes of the basic entities. The attributes of the basic entities include the key entity identifiers to which the basic entities belong, and the key entity to which a basic entity belongs is unique.

[0038] Exemplarily, extract the keyword fields of the target user from the historical service data of the target user, calculate the correlation between the keyword fields of the target user and the basic entity fields in the knowledge graph in the resource allocation scenario, and obtain the successfully matched basic entities according to the correlation; construct an initial network according to the key entities to which the basic entities belong and the edge relationships between different basic entities, and different types of edge relationships and different key entities can be distinguished and displayed. Refer to Figure 1b, the initial network may include a first key entity, a second key entity, a third key entity, and a fourth key entity. The first key entity may include four different basic entities, the second key entity may include two different basic entities, the third key entity may include three different key entities, and the fourth key entity may include two different basic entities. Moreover, the corresponding edge relationships can be distinguished and shown using different line segments (such as line segments of different colors). Among them, a key entity includes its affiliated basic entities in the form of a container, that is, the spatial nesting between the basic entity and the key entity to which it belongs is used to represent the hierarchical subordination relationship between the basic entity and the key entity. 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 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, and the collection, use, and processing of relevant data comply with the relevant laws, regulations, and standards of the relevant region.

[0039] By semantically matching the historical business data of the target user with the knowledge graph of resource allocation, various basic entities related to resource allocation can be accurately extracted. Combining the edge relationships between different basic entities and the key entities to which each basic entity belongs, a visual initial network is constructed, enabling the topological structure of the initial network to present the interaction logic of multiple types of constraints in resource allocation, laying a foundation for the parametric modeling of subsequent distributed constraint optimization algorithms. Moreover, by introducing the relationship between the basic entity and the key entity, it is convenient to dynamically adjust the basic entity through the key entity, thereby improving the adjustment efficiency and accuracy of distributed constraint optimization.

[0040] The user can perform addition or deletion operations on the basic entities in the initial network, and can also edit the edge relationships, such as adjusting the type of the edge relationship or modifying the constraint parameters in the edge relationship. Exemplarily, in the case where there is an edge relationship of optimal ratio constraint between two basic entities, the corresponding optimal ratio can be adjusted from the first ratio value to the second ratio value; in the case where there is an inverse constraint edge relationship between two basic entities, the ratio of one basic entity can be decreased while the ratio of the other basic entity is increased. By updating the basic entities and edge relationships in the initial network in response to the editing operation on the initial network to obtain the target network, the matching degree between the target network and the user's resource allocation requirements can be further improved, reducing the number of adjustments in resource allocation, thereby improving the resource allocation efficiency.

[0041] The distributed constraint optimization problem is a commonly used technology for solving distributed system modeling and collaborative optimization problems. In the embodiments of the present application, through the interaction between agents, high parallelism and fault tolerance are achieved, and the global optimum can be reached. Exemplarily, agents and variables in the agents can be determined based on the key entities to which the basic entities in the target network belong; for each agent, the local constraint cost of the agent can be constructed according to the edge relationship between different basic entities within the agent; and the global constraint cost can be constructed according to the cross-agent edge relationship to obtain a distributed constraint optimization model. In the solving process, each agent not only independently calculates the local constraint cost and updates the local variables of the corresponding agent according to the local constraint cost, but also different agents communicate to exchange variable state information and constraints in real time, and dynamically adjust the neighbor variable values based on the global consistency protocol. By minimizing the local constraint cost of each agent and the global constraint cost across agents, the values of each basic variable are obtained, and the resource allocation scheme for the target user is obtained according to the values of each basic variable.

[0042] The technical solution provided in the embodiments of the present application performs semantic matching on the historical service data of the target user and the knowledge graph of resource allocation, extracts the basic entities related to resource constraints, and combines the edge relationship between the basic entities and the key entities to which the basic entities belong to construct a visual initial network, so that the initial network can present the interaction logic of various types of resource constraints, which not only lays a foundation for the parametric modeling of the subsequent distributed constraint optimization algorithm. And, in response to the interactive editing operation on the initial network, the entity nodes and edge relationships are updated in real time to form a target network, further improving the matching degree between the target network and the actual needs of the user, and effectively reducing the number of scheme iterations. Moreover, through the dual mechanisms of each agent independently calculating the local constraint cost and collaboratively optimizing the global constraint cost, the optimal values of each basic variable are efficiently determined under the distributed solving framework, so as to generate a resource allocation scheme with both theoretical optimality and practical feasibility, improving the accuracy and efficiency of resource allocation.

[0043] Embodiment 2

[0044] Figure 2 is a flowchart of the resource allocation method based on distributed constraint optimization provided in Embodiment 2 of the present application. Refer to Figure 2 , on the basis of the above embodiments, the resource allocation method based on distributed constraint optimization in this embodiment may include:

[0045] S210. Match the historical service data of the target user with the knowledge graph of resource allocation to obtain at least two basic entities, and construct and display an initial network based on the key entities to which the basic entities belong and the edge relationship between different basic entities;

[0046] S220. Obtain the real-time business data of the business system, and generate at least two candidate adjustment strategies for resource allocation according to the real-time business data;

[0047] S230. Match the candidate adjustment strategies with the initial network respectively, and select a target adjustment strategy from the at least two candidate adjustment strategies according to the matching results;

[0048] S240. Interact with the user according to the target adjustment strategy, and edit the initial network according to the user interaction operation to obtain a target network;

[0049] S250. Construct a distributed constraint optimization model according to the target network and the historical business data;

[0050] S260. Solve the distributed constraint optimization model to obtain a resource allocation plan for the target user.

[0051] Among them, the historical business data of the target user refers to static information such as the behavior records, preference characteristics, and transaction history accumulated by the user for a long time, which is used to characterize the individual behavior pattern and stable characteristics of the target user; while the real-time business data of the business system comes from the external environment or the global operation system outside the target user, such as dynamic information such as market volatility indicators, policy change signals, and emergency warnings, which is used to characterize the real-time situation of the business field and provide a dynamic adjustment basis for resource allocation strategies to cope with sudden changes in industry trends or crisis events.

[0052] Exemplarily, perform semantic analysis on the real-time business data of the business system to determine whether there are market volatility indicators or policy change signals, etc., to obtain the target dynamic information existing in the business system. If "reserve requirement ratio cut", "interest rate cut", "industry correction", "the market index returns to... points", etc. appear in the real-time business data, the market volatility indicator is triggered; if "tax incentives", "subsidies", etc. appear in the real-time business data, the policy change signal is triggered, etc. Match the target dynamic information with the policy rules in the rule library to obtain at least two candidate adjustment strategies with successful matches.

[0053] For each candidate adjustment strategy, the node coverage rate of the candidate adjustment strategy for the initial network can be determined, and the candidate adjustment strategy with the largest node coverage rate is used as the target adjustment strategy. Exemplarily, the candidate adjustment strategy can be semantically encoded to obtain a semantic encoding result, and the semantic encoding result is matched with the basic entities, key entities or edge relationships in the initial network to obtain the matching degree of the candidate adjustment strategy; where the matching degree is used to represent the node coverage rate. Moreover, candidate edit operations are generated using the target adjustment strategy, and the target edit operation is obtained by interacting with the user based on the candidate edit operations. The initial network is updated using the target edit operation to obtain the target network, that is, the initial network is updated based on the target adjustment strategy to obtain the target network. By processing the real-time business data of the business system, at least two candidate adjustment strategies are obtained. The node coverage rates of the candidate adjustment strategies for the initial network are determined respectively, and the candidate adjustment strategy with the largest node coverage rate is used as the target adjustment strategy. On the one hand, it can simplify the editing operation of the initial network and improve the determination efficiency of the target network; on the other hand, the target network not only contains the individual behavior patterns of the target user, but also can take into account the dynamics of industry trends, thereby improving the reliability of the target network.

[0054] In an alternative embodiment, the step of matching the candidate adjustment strategies with the initial network respectively and selecting a target adjustment strategy from the at least two candidate adjustment strategies according to the matching results includes: matching the candidate adjustment strategies with the initial network respectively to obtain candidate sub-networks that match successfully; determining the importance of the candidate sub-networks according to the number of basic entities and the number of constraint edges in the candidate sub-networks, and the historical editing times of the target user for the candidate sub-networks; and selecting a target adjustment strategy from the at least two candidate adjustment strategies according to the importance.

[0055] For each candidate adjustment strategy, the candidate adjustment strategy can be semantically encoded to obtain the semantic encoding result of the candidate adjustment strategy; calculate the similarity between the semantic encoding result of the candidate adjustment strategy and the encoding results of each basic entity and the edge relationship in the initial network respectively, and use the basic entities and edge relationships with similarity greater than the similarity threshold to obtain a candidate sub-network; combine the number of basic entities, the number of constraint edges, and the historical editing times of the target user on the candidate sub-network to determine the importance of the candidate sub-network. Exemplarily, the importance of the candidate sub-network can be obtained by weighting the number of basic entities, the number of constraint edges, and the historical editing times, and the candidate adjustment strategy corresponding to the candidate sub-network with the highest importance is used as the target adjustment strategy. By semantically encoding and calculating the similarity of the candidate adjustment strategy, the candidate sub-network corresponding to the candidate adjustment strategy can be accurately screened out, and the importance of the candidate sub-network is determined by combining the number of basic entities, the number of constraint edges, and the historical editing times of the candidate sub-network, which can efficiently and accurately determine the target adjustment strategy, thereby further improving the efficiency and accuracy of resource allocation.

[0056] In an alternative embodiment, the step of interacting with the user according to the target adjustment strategy and editing the initial network according to the user interaction operation to obtain the target network includes: determining a first entity associated with the target adjustment strategy from the initial network, and obtaining a second entity having an edge relationship with the first entity; generating and displaying a candidate editing operation according to the edge relationship between the first entity and the second entity; wherein the edge relationship includes at least one of the following: optimal ratio constraint, reverse constraint, or conditional optimal configuration constraint; updating the constraint relationship between the first entity and the second entity according to the interaction processing of the candidate editing operation to obtain the target network.

[0057] Exemplarily, obtain the semantic encoding result of the target adjustment strategy by semantically encoding the target adjustment strategy; calculate the similarity between the semantic encoding result of the target adjustment strategy and the encoding results of each basic entity in the initial network respectively, and use the basic entity with the largest similarity as the first entity; also obtain other basic entities connected to the first entity from the initial network as the second entity; generate and display a candidate editing operation based on the edge relationship between the first entity and the second entity and according to the type of the edge relationship, and update the constraint relationship between the first entity and the second entity in response to the interaction operation on the candidate editing operation to obtain the target network.

[0058] Among them, the edge relationship between the first entity and the second entity includes but is not limited to an optimal ratio constraint, a reverse constraint, or a conditional optimal configuration constraint. If the edge relationship between the two is an optimal ratio constraint, an optimal configuration adjustment control can be displayed in the interaction interface to respond to the user operation to move the position of the control, thereby adjusting the ratio between the first entity and the second entity. If the edge relationship between the two is a reverse constraint, an adjustment control for the first entity can be displayed in the interaction interface for the user to increase or decrease the quantity of the first entity. If the edge relationship between the two is a conditional optimal configuration constraint, a conditional input box can be displayed in the interaction interface for the user to define conditions such as the time range, resource threshold, or priority for which the constraint becomes effective. By using the basic entity in the initial network that is most similar to the target adjustment strategy result as the first entity, and providing candidate editing operations according to the constraint type between the first entity and the second entity for the target user to update the constraint relationship between the two to obtain the target network, the efficiency and accuracy of network adjustment are further improved.

[0059] The technical solution provided by the embodiments of the present application processes the real-time service data of the service system to obtain at least two candidate adjustment strategies, determines the node coverage rate of each candidate adjustment strategy for the initial network, and selects the one with the largest coverage rate as the target adjustment strategy. This not only simplifies the editing operation of the initial network, improves the determination efficiency of the target network, but also enables the target network to incorporate industry trend dynamics and enhances reliability. In addition, by matching the candidate adjustment strategy with the initial network to obtain a candidate sub-network, and determining the target adjustment strategy in combination with the number of basic entities, the number of constraint edges, and the historical editing times of the candidate sub-network; setting the basic entity in the initial network that is most similar to the target adjustment strategy as the first entity, and providing candidate editing operations according to the constraint type for the target user to update the constraint relationship, thereby further optimizing the efficiency and accuracy of network adjustment.

[0060] Figure 3 It is a flowchart of a resource allocation method based on distributed constraint optimization provided by Embodiment 2 of the present application. Refer to Figure 3 On the basis of the above embodiments, the resource allocation method based on distributed constraint optimization in this embodiment may include:

[0061] S310. Match the historical service data of the target user with the knowledge graph of resource allocation to obtain at least two basic entities, and construct and display an initial network based on the key entities to which the basic entities belong and the edge relationships between different basic entities;

[0062] S320. Obtain a target network in response to an editing operation on the initial network;

[0063] S330. Take each type of key entity in the target network as an agent, take the basic entities connected by the key entity as variables corresponding to the agent, and determine the target constraint relationship between different variables according to the edge relationship between different basic entities;

[0064] S340. Extract the value range of the variable and the parameter value of the target constraint relationship from the historical service data, and determine the target constraint cost function between different variables according to the parameter value;

[0065] S350. Build the distributed constraint optimization model based on the agent, the variable, the value range, and the target constraint cost function;

[0066] S360. Solve the distributed constraint optimization model to obtain the resource allocation scheme for the target user.

[0067] In the embodiment of the present invention, the distributed constraint model can be represented as a quadruple <A, X, D, F> model. Among them, A represents the set of agents. Each type of key entity in the target network can be taken as an agent. Taking four key entities as an example, four agents are correspondingly constructed. X represents the set of variables. Each agent controls multiple variables. The basic entities connected by the key entity are taken as the variables corresponding to the agent. Taking a certain key entity connected to four basic entities as an example, the corresponding agent includes four variables. D represents the set of value ranges. Each variable takes values from its value range. F represents the set of constraint costs. The local constraint cost inside the agent is constructed according to the edge relationship between different basic entities in the same agent; the global constraint cost across agents is constructed according to the edge relationship between different basic entities across agents. That is, the set of constraint costs includes local constraint costs and global constraint costs.

[0068] The value range of the variable and the parameter value of the target constraint relationship can both be extracted from the historical service data of the target user. The target constraint relationship may include at least one of the following: optimal ratio constraint, reverse constraint, or conditional optimal configuration constraint. Exemplarily, when the ratio of the first variable is the first value a and the ratio of the second variable is the second value b, the cost function z = (x - a1) 2 +(y - b1) 2Reaches the minimum value, that is, the lowest point is (a1, b1). Then the constraint relationship between the first variable and the second variable can be described as an optimal ratio constraint. If as the value of the second variable increases, the proportion of the first variable gradually decreases, the constraint cost between the two is defined by the formula -a2x - b2y + c2 = z, where a2, b2, and c2 are constants. In this case, the constraint relationship between the first variable and the second variable can be called an inverse ratio constraint or a negative correlation constraint. Exemplarily, in terms of the properties of gold, it belongs to a hedging resource. The farther away from the retirement time, the lower the proportion of the gold variable. Therefore, there is an inverse constraint between the environmental variable and the retirement time variable.

[0069] Conditional optimal allocation constraint refers to the definition or rule of finding the optimal allocation plan that satisfies specific constraint conditions by adjusting the allocation variables to minimize or maximize a certain objective function under given conditions. Exemplarily, according to the type of customer (such as growth type) and the evaluation coefficient y (normalized to [1, 10]), the golden ratio allocation is carried out to obtain the conditional optimal allocation constraint relationship z = y × |x - I|, where x ∈ [1, 10], y ∈ [1, 10], and I is the extreme value corresponding to minimization or maximization. Taking the constraint relationship between the estate planning variable and the gold variable as an example, if the optimal gold ratio allocation for the target user is 5%, and the value of I obtained by normalizing 5% is 1.45, the constraint relationship z = y × |x - 1.45| can be obtained, where x ∈ [1, 10], y ∈ [1, 10].

[0070] Taking the constraint relationship between the insurance purchase decision and the estate planning variable as an example, the optimal insurance purchase demand of the target user can be calculated by the following formula:

[0071]

[0072] where I is the optimal insurance purchase demand, N is the estate planning demand [0, 1 million]; R is the risk tolerance level evaluation coefficient, obtained from the business system; V0 is the actual estate value; V A is the estate planning target value. Since the value ranges of different variables are different, the value ranges of each variable can be normalized. Taking the retirement time variable as an example, its value is [0, 40], and the value range can be updated in real time according to the retirement policy. The following normalization formula can be used to normalize the retirement time to integers from 1, 2,.., 10:

[0073]

[0074] Moreover, distributed constraint optimization algorithms (such as ant colony algorithms, distributed constraint satisfaction algorithms, etc.) can be used for solving. When the local constraint cost and the global constraint cost are minimized, the variable values of each agent are obtained, that is, according to different investment preferences and risk tolerance levels, combined with real-time dynamic information, the variable values of various candidate resources are obtained as the resource allocation scheme for the target user. Through relevant platform interfaces, real-time personalized recommendations and displays of the resource allocation scheme can be carried out.

[0075] In the technical solution provided by the embodiment of the present application, by taking the key entities in the target network as agents and the basic entities connecting the key entities as variables, by determining the target constraint relationship between the variables, the value range of the variables and the parameter values of the constraints are extracted from the historical service data of the target user, and a distributed constraint optimization model is constructed accordingly, which can flexibly respond to complex and changeable service requirements and improve the resource allocation efficiency.

[0076] In an alternative embodiment, the initial network includes at least two of the following key entities: resource allocation type, time decision type, risk management decision type or inheritance decision type; the key entity of the resource allocation type includes at least two basic entities of candidate resources; the key entity of the time decision type includes retirement time or investment period; the key entity of the risk management decision type includes at least one of insurance purchase decision, medical expense budget or risk tolerance level; the key entity of the inheritance decision type includes estate planning or grandchild education fund.

[0077] Exemplarily, the distributed constraint optimization model may include an agent of the resource allocation type, and also includes at least one of an agent of the time decision type, an agent of the risk management decision type or an agent of the inheritance decision type; the agent of the resource allocation type may include at least one of the following candidate resource variables: stock investment, bond investment, wealth management product investment, derivatives such as gold; the agent of the time decision type may include a retirement time or investment period variable; the agent of the risk management decision type may include at least one variable of insurance purchase decision, medical expense budget or risk tolerance level; the agent of the inheritance decision type may include an estate planning or grandchild education fund variable. By constructing and solving the distributed constraint optimization model, the value of each candidate variable can be obtained as the resource allocation scheme for the target user.

[0078] Embodiment 4

[0079] Figure 4 FIG. 16 is a schematic structural diagram of a resource allocation device based on distributed constraint optimization according to Embodiment 4 of the present application, which is applicable to the situation of allocating user resources, and can be executed by a resource allocation device based on distributed constraint optimization. The resource allocation device based on distributed constraint optimization can be implemented in the form of hardware and / or software, and the resource allocation device based on distributed constraint optimization can be configured in a computer device. As Figure 4As shown, the device includes:

[0080] An initial network module 410, configured to match the historical service data of a target user with a knowledge graph of resource allocation to obtain at least two basic entities, and construct and display an initial network based on the key entities to which the basic entities belong and the edge relationships between different basic entities;

[0081] A target network module 420, configured to obtain a target network in response to an editing operation on the initial network;

[0082] A distributed constraint optimization module 430, configured to construct a distributed constraint optimization model according to the target network and the historical service data;

[0083] A resource allocation scheme module 440, configured to solve the distributed constraint optimization model to obtain a resource allocation scheme for the target user.

[0084] In an alternative embodiment, the target network module 420 includes:

[0085] A candidate policy sub-module, configured to obtain real-time service data of a service system and generate at least two candidate adjustment policies for resource allocation according to the real-time service data;

[0086] A target policy sub-module, configured to respectively match the candidate adjustment policies with the initial network, and select a target adjustment policy from the at least two candidate adjustment policies according to the matching results;

[0087] A target network sub-module, configured to interact with a user according to the target adjustment policy, and edit the initial network according to the user interaction operation to obtain a target network.

[0088] In an alternative embodiment, the target policy sub-module includes:

[0089] A candidate sub-network unit, configured to respectively match the candidate adjustment policies with the initial network to obtain candidate sub-networks with successful matches;

[0090] An importance unit, configured to determine the importance of a candidate sub-network according to the number of basic entities and the number of constraint edges in the candidate sub-network, and the historical editing times of the target user for the candidate sub-network;

[0091] A target policy unit, configured to select a target adjustment policy from the at least two candidate adjustment policies according to the importance.

[0092] In an alternative embodiment, the target network sub-module includes:

[0093] An entity determination unit, configured to determine a first entity associated with the target adjustment policy from the initial network, and obtain a second entity having an edge relationship with the first entity;

[0094] An editing operation unit, configured to generate and display candidate editing operations according to the edge relationship between the first entity and the second entity; wherein the edge relationship includes at least one of the following: optimal ratio constraint, reverse constraint, or conditional optimal configuration constraint;

[0095] A target network unit, configured to update the constraint relationship between the first entity and the second entity according to the interaction processing of the candidate editing operations, and obtain a target network.

[0096] In an alternative embodiment, the distributed constraint optimization module 430 includes:

[0097] A model conversion sub-module, configured to use each key entity in the target network as an agent, use the basic entities connected to the key entity as variables corresponding to the agent, and determine the target constraint relationship between different variables according to the edge relationship between different basic entities;

[0098] A parameter extraction sub-module, configured to extract the value range of the variable and the parameter value of the target constraint relationship from the historical service data, and determine the target constraint cost function between different variables according to the parameter value;

[0099] A model construction sub-module, configured to construct the distributed constraint optimization model based on the agent, the variable, the value range, and the target constraint cost function.

[0100] In an alternative embodiment, the initial network includes at least two of the following key entities: resource configuration type, time decision type, risk management decision type, or inheritance decision type; the key entity of the resource configuration type includes basic entities of at least two candidate resources; the key entity of the time decision type includes retirement time or investment period; the key entity of the risk management decision type includes at least one of insurance purchase decision, medical expense budget, or risk tolerance level; the key entity of the inheritance decision type includes estate planning or grandchild education fund.

[0101] The resource allocation device based on distributed constraint optimization provided by the embodiments of the present application can execute the resource allocation method based on distributed constraint optimization provided by any embodiment of the present application, and has corresponding functional modules and beneficial effects for executing each resource allocation method based on distributed constraint optimization.

[0102] According to the embodiments of the present application, the present application also provides an electronic device, a readable storage medium, and a computer program product.

[0103] Embodiment 5

[0104] Figure 5 FIG. 510 is a schematic structural diagram of an electronic device 510 for implementing the resource allocation method based on distributed constraint optimization according to the embodiments of the present application. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, 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 illustrative and are not intended to limit the implementation of the present application described and / or claimed herein.

[0105] As Figure 5 shown, the electronic device 510 includes at least one processor 511, and a memory communicatively connected to the at least one processor 511, such as a read-only memory (ROM) 512, a random access memory (RAM) 513, etc. The memory stores a computer program executable by the at least one processor. The processor 511 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 512 or the computer program loaded from the storage unit 518 into the random access memory (RAM) 513. In the RAM 513, various programs and data required for the operation of the electronic device 510 can also be stored. The processor 511, the ROM 512, and the RAM 513 are connected to each other through a bus 514. An input / output (I / O) interface 515 is also connected to the bus 514.

[0106] Multiple components in the electronic device 510 are connected to the I / O interface 515, including: an input unit 516, such as a keyboard, a mouse, etc.; an output unit 517, such as various types of displays, speakers, etc.; a storage unit 518, such as a magnetic disk, an optical disk, etc.; and a communication unit 519, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 519 allows the electronic device 510 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0107] The processor 511 may be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 511 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 511 executes the various methods and processes described above, such as the resource allocation method based on distributed constraint optimization.

[0108] In some embodiments, the resource allocation method based on distributed constraint optimization can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 518. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 510 via the ROM 512 and / or the communication unit 519. When the computer program is loaded into the RAM 513 and executed by the processor 511, one or more steps of the resource allocation method based on distributed constraint optimization described above can be executed. Alternatively, in other embodiments, the processor 511 can be configured to be the resource allocation method based on distributed constraint optimization by any other suitable means (e.g., by means of firmware).

[0109] Various embodiments of the systems and techniques described above in this document can 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), systems-on-a-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0110] The computer programs for implementing the methods of this application can be written in any combination of one or more programming languages. These computer programs can be provided to the processors of general-purpose computers, special-purpose computers, or other programmable devices such that when the computer programs are executed by the processors, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer programs can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0111] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0112] To provide for interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide for interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0113] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0114] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0115] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in this application can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of this application can be achieved, and no limitation is made herein.

[0116] The above specific embodiments do not constitute a limitation on the protection scope of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the protection scope of this application.

Claims

1. A resource allocation method based on distributed constraint optimization, characterized in that Including: Matching the historical business data of the target user with the knowledge graph of resource allocation to obtain at least two basic entities, and constructing and displaying an initial network based on the key entities to which the basic entities belong and the edge relationships between different basic entities; Obtaining a target network in response to an editing operation on the initial network; Constructing a distributed constraint optimization model according to the target network and the historical business data; Solving the distributed constraint optimization model to obtain a resource allocation scheme for the target user.

2. The method according to claim 1, characterized in that, The obtaining a target network in response to an editing operation on the initial network includes: Obtaining the real-time business data of the business system, and generating at least two candidate adjustment strategies for resource allocation according to the real-time business data; Matching the candidate adjustment strategies with the initial network respectively, and selecting a target adjustment strategy from the at least two candidate adjustment strategies according to the matching results; Interacting with the user according to the target adjustment strategy, and editing the initial network according to the user interaction operation to obtain a target network.

3. The method according to claim 2, wherein, The matching the candidate adjustment strategies with the initial network respectively, and selecting a target adjustment strategy from the at least two candidate adjustment strategies according to the matching results includes: Matching the candidate adjustment strategies with the initial network respectively to obtain candidate sub-networks with successful matches; Determining the importance of the candidate sub-networks according to the number of basic entities and the number of constraint edges in the candidate sub-networks, and the historical editing times of the candidate sub-networks by the target user; Selecting a target adjustment strategy from the at least two candidate adjustment strategies according to the importance.

4. The method according to claim 2, characterized in that The interacting with the user according to the target adjustment strategy, and editing the initial network according to the user interaction operation to obtain a target network includes: Determining a first entity associated with the target adjustment strategy from the initial network, and obtaining a second entity having an edge relationship with the first entity; Generating and displaying candidate editing operations according to the edge relationship between the first entity and the second entity; wherein, the edge relationship includes at least one of the following: optimal ratio constraint, reverse constraint or conditional optimal configuration constraint; Updating the constraint relationship between the first entity and the second entity according to the interactive processing of the candidate editing operations to obtain a target network.

5. The method according to claim 1, characterized in that The constructing a distributed constraint optimization model according to the historical business data based on the target network includes: Regarding each type of key entity in the target network as an agent, regarding the basic entities connected by the key entity as variables corresponding to the agent, and determining the target constraint relationship between different variables according to the edge relationships between different basic entities; Extracting the value range of the variables and the parameter values of the target constraint relationship from the historical business data, and determining the target constraint cost function between different variables according to the parameter values; Constructing the distributed constraint optimization model based on the agent, the variable, the value range and the target constraint cost function.

6. The method according to claim 1, characterized in that, The initial network includes at least two of the following key entities: resource allocation class, time decision class, risk management decision class, or inheritance decision class; the key entity of the resource allocation class includes the basic entities of at least two candidate resources; the key entity of the time decision class includes retirement time or investment period; the key entity of the risk management decision class includes at least one of insurance purchase decision, medical expense budget, or risk tolerance level; the key entity of the inheritance decision class includes estate planning or education fund for grandchildren.

7. A resource allocation device based on distributed constraint optimization, characterized in that including: An initial network module, configured to match the historical business data of the target user with the knowledge graph of resource allocation to obtain at least two basic entities, and construct and display an initial network based on the key entities to which the basic entities belong and the edge relationships between different basic entities; A target network module, configured to obtain a target network in response to an editing operation on the initial network; A distributed constraint optimization module, configured to construct a distributed constraint optimization model according to the target network and the historical business data; A resource allocation scheme module, configured to solve the distributed constraint optimization model to obtain a resource allocation scheme for the target user.

8. An electronic device, characterized in that, including: One or more processors; A memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1-6.

10. A computer program product, including a computer program, where the computer program implements the method according to any one of claims 1-6 when executed by a processor.

Images