Service resource configuration method and device, equipment and storage medium

Abstract

The invention discloses a service resource configuration method and device, equipment and a storage medium. The method comprises the following steps: splicing a state cluster label of a target cluster with feature information to obtain updated feature information; inputting the updated feature information into an abnormal risk assessment model corresponding to the target cluster to obtain an abnormal probability; determining a risk level adjustment coefficient of the current service according to the state cluster label, and determining a resource configuration parameter of the current service according to the abnormal probability and the risk level adjustment coefficient; and performing resource configuration on the current service according to the resource configuration parameter. The features of the current service are updated through the state cluster label of the matched target cluster so as to enrich the feature information of the current service, and the abnormal probability is directly acquired by adopting a pre-trained abnormal risk assessment model based on the updated feature information; and the resource configuration parameters of the current service are comprehensively and accurately determined based on two factors of the abnormal probability and the risk level adjustment coefficient determined by the state cluster label.

Description

Technical Field

[0001] This invention relates to the field of financial technology, and more particularly to a method, apparatus, device, and storage medium for resource allocation in business operations. Background Technology

[0002] Bank systems typically execute a large number of transactions every day to meet the diverse needs of customers. For customers with similar needs, transactions are usually processed in batches, and resources are allocated in a fixed manner for each batch to improve efficiency.

[0003] Because different business operations handle transactions of varying difficulty, there may be an overabundance of resources for simple transactions and an underabundance of resources for complex transactions, which may even lead to business interruptions. This can affect the normal operation of the overall business, fail to meet customer needs, and reduce customer service experience. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and storage medium for configuring business resources, so as to achieve accurate and efficient configuration of business resources.

[0005] According to a first aspect of the present invention, a resource allocation method for a service is provided, the method comprising: determining a matching target cluster from a cluster set based on feature information of the current service, and concatenating the state cluster label of the target cluster with the feature information to obtain updated feature information; The updated feature information is input into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability; The risk level adjustment coefficient of the current service is determined based on the state cluster label, and the resource configuration parameters of the current service are determined based on the anomaly probability and the risk level adjustment coefficient. Configure resources for the current service according to the resource configuration parameters.

[0006] According to another aspect of the present invention, a resource allocation apparatus for a business is provided, the apparatus comprising: The updated feature information acquisition module is used to determine the matching target cluster from the cluster set based on the feature information of the current business, and to concatenate the state cluster label of the target cluster with the feature information to obtain updated feature information; An anomaly probability assessment module is used to input the updated feature information into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability; The resource configuration parameter determination module is used to determine the risk level adjustment coefficient of the current service based on the status cluster label, and to determine the resource configuration parameters of the current service based on the anomaly probability and the risk level adjustment coefficient. The resource configuration module is used to configure resources for the current service according to the resource configuration parameters.

[0007] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: one or more processors; Storage device for storing 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 described in any embodiment of the present invention.

[0008] According to another aspect of the present invention, a storage medium for computer-executable instructions is provided, on which a computer program is stored, which, when executed by a processor, implements the method as described in any of the embodiments of the present invention.

[0009] The technical solution of this invention updates the features of the current business by matching the state cluster labels of the target clusters to enrich the feature information of the current business. Based on the updated feature information, a pre-trained anomaly risk assessment model is used to directly obtain the anomaly probability, thereby quickly identifying the operation status of the current business. Based on the anomaly probability and the risk level adjustment coefficient determined by the state cluster labels, the resource configuration parameters of the current business are comprehensively and accurately determined, so as to achieve accurate and efficient allocation of business resources according to the resource configuration parameters.

[0010] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart of a service resource allocation method provided according to Embodiment 1 of the present invention; Figure 2 This is a flowchart of another service resource configuration method provided according to Embodiment 2 of the present invention; Figure 3This is a schematic diagram of the structure of a service resource allocation device provided according to Embodiment 3 of the present invention; Figure 4 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0013] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0014] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or terminal device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or terminal devices.

[0015] Example 1 Figure 1 This is a flowchart of a service resource configuration method provided in Embodiment 1 of the present invention. This embodiment is applicable to the configuration of service resources. The method can be executed by a service resource configuration device, which can be implemented in hardware and / or software, and can be integrated into an electronic device with data processing capabilities. Figure 1 As shown, the method includes: S101, based on the feature information of the current business, determine the matching target cluster from the cluster set, and concatenate the state cluster label of the target cluster with the feature information to obtain the updated feature information.

[0016] Optionally, before determining the matching target cluster from the cluster set based on the feature information of the current business, the method further includes: obtaining the feature information of the sample business, and clustering the sample business according to the feature information of the sample business to obtain multiple clusters; performing state recognition on each cluster to obtain the state of the cluster, and adding state cluster labels to the clusters according to the state; and constructing a cluster set based on the clusters with added state cluster labels.

[0017] Specifically, in this embodiment, before allocating resources for the current business, historical business transactions that have already been executed are retrieved from the financial system database. These retrieved historical transactions are used as sample transactions. By clustering these sample transactions, the execution patterns of business transactions in the financial system are obtained and applied to the resource allocation process of the current business. For example, this embodiment obtains the characteristic information of the sample transactions. This characteristic information may specifically include time-series features containing multiple indicators. Of course, this embodiment is only an example and does not limit the specific content or form of the characteristic information. In addition, the sample transactions can be clustered based on their characteristic information to obtain multiple clusters. Furthermore, to eliminate the influence of amplitude and make the clustering process focus more on morphological similarity, the characteristic information can be standardized to improve the efficiency and accuracy of clustering.

[0018] In this embodiment, when clustering sample services, the similarity between each sample service can be calculated. For example, the similarity can be determined based on the distance between feature information, and then the sample services can be clustered based on the similarity to obtain multiple clusters. Each cluster contains multiple sample services with the same running status. Of course, this embodiment is only an example and does not limit the specific calculation process of similarity.

[0019] In this embodiment, after obtaining multiple clusters through clustering, the state of each cluster is identified. For example, the state of each sample service in the cluster is identified, and the state with the largest proportion is taken as the running state of the cluster. The identified state is added as a state cluster label to the cluster. A cluster set is constructed based on the clusters with added state cluster labels. For example, the cluster set C={C1,C2...C10}, that is, the cluster set contains 10 clusters. Of course, this embodiment is only an example and does not limit the number of clusters contained in the cluster set.

[0020] Optionally, a matching target cluster can be determined from the cluster set based on the feature information of the current business, including: calculating the distance between the feature information and the center of each cluster in the cluster set; and taking the cluster with the smallest distance as the target cluster that matches the current business.

[0021] In this embodiment, after clustering the sample services to obtain a set of clusters containing multiple clusters, for any current service in the financial system, the matching target cluster is queried from the set of clusters based on the collected feature information. When matching the target cluster, the distance between the feature information of the current service and the center of each cluster can be calculated. The smaller the distance, the more similar the current service is to the sample services in that cluster. Therefore, the cluster with the smallest distance is selected as the target cluster matching the current service. Since the current service state is similar to that of the sample services in the target cluster, the state cluster label of the target cluster can be directly used as the state parameter of the current service. The state cluster label is then concatenated with the feature information to obtain the updated feature information, which includes the state information of the current service. Therefore, in this embodiment, there is no need to perform state recognition for the current service. It is only necessary to match it with the clusters whose states have already been identified, and obtain the state information of the current service based on the state cluster label of the matched target cluster. Furthermore, the algorithm complexity of the matching process is significantly reduced compared to the recognition process. Thus, in this embodiment, the state recognition of the current service is achieved through the pre-constructed clusters, enriching the feature information of the current service.

[0022] S102, update the feature information and input it into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability.

[0023] Optionally, before inputting the updated feature information into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability, the method further includes: obtaining a corresponding subset for each cluster, using an improved sparrow algorithm to obtain the optimal hyperparameter combination corresponding to the subset; configuring the gradient boosting decision tree framework according to the optimal hyperparameter combination, and using the subset to train the configured gradient boosting decision tree framework to obtain the anomaly risk assessment model corresponding to the cluster.

[0024] Optionally, an improved sparrow algorithm is used to obtain the optimal hyperparameter combination corresponding to the subset, including: randomly generating an initial population of a specified size for each subset; calculating the fitness of individuals and sorting them to determine the discoverer, the watcher, and the follower, as well as the random step size of the follower; updating the positions of the discoverer and the watcher, and updating the positions of the follower based on the random step size; evaluating the individuals after the position update, and taking the obtained global optimal solution as the optimal hyperparameter combination when the iteration condition is met.

[0025] Specifically, in this embodiment, a corresponding anomaly risk assessment model is created for each cluster. During model creation, the feature information of the sample services included in each cluster is concatenated with state labels and binary anomaly labels to obtain the final feature vector. Based on the final feature vector of each sample service, a subset of data points corresponding to the cluster is constructed. Each subset of data points is then trained to obtain the anomaly risk assessment model corresponding to that cluster. Since different clusters have significant differences in risk formation mechanisms and feature distributions, this embodiment optimizes each subset of data points to obtain the optimal hyperparameter combination for model training. This effectively improves the predictive performance and stability of the anomaly risk assessment model corresponding to each cluster. Specifically, this embodiment uses an improved sparrow algorithm for optimization to obtain the optimal hyperparameter combination corresponding to each subset of data points. .

[0026] For example, the problem to be solved in this implementation is to find the optimal combination of hyperparameters for a gradient boosting decision tree framework for a subset D1 (obtained from cluster C1). To maximize cross-validation performance on subset D1, the fitness function is... The specific process for optimization using the improved sparrow algorithm is as follows: First, randomly generate an initial population of a specified size for each subset of data, such as... Second, calculate and rank the individual fitness to determine the discoverer, watchdog, and follower, as well as the random step size of the follower; Third, update the position of the discoverer using the following formula (1): in, During the nth iteration Only sparrows in the first The position of the dimension It is a safety threshold. Additionally, the position of the follower is updated using the following formula (2): in, This is the current optimal discoverer position. This is the step size control factor. The step size is random. The step size is a control parameter, usually taken as 1.5, and the random step size can be calculated using the following formula (3): in, In addition, the position of the vigilant is updated using the following formula (3): in, It is the current globally optimal position. These are the fitness values ​​of the current individual, the globally best individual, and the worst individual, respectively. It is a step size control parameter. It is a random number. This is a minimal constant to avoid division by zero errors. Fourth, this implementation method will continuously repeat steps two and three until the stopping condition is met, thereby finding the corresponding optimal hyperparameter combination for the subset D1. Of course, this embodiment only illustrates the method of obtaining the optimal hyperparameter combination corresponding to the subset D1. The method of obtaining the optimal hyperparameter combination corresponding to other subsets is roughly the same, and will not be elaborated upon in this embodiment. In addition, by introducing a random step size when updating the discoverer's position, this embodiment can effectively avoid getting trapped in local optima and facilitate finding the optimal solution quickly and efficiently.

[0027] It is worth mentioning that, in this embodiment, after obtaining the optimal hyperparameter combination corresponding to each subset of data through optimization, the optimal hyperparameter combination is used to configure the gradient boosting decision tree framework, and the configured framework is trained using the subset of data to obtain the anomaly risk assessment model corresponding to each cluster. For example, the anomaly risk assessment model M1 corresponds to cluster C1. This embodiment does not limit the number of anomaly risk assessment models obtained; as long as the number is the same as the number of clusters, it is within the scope of protection of this application. In addition, after obtaining the anomaly risk assessment model corresponding to each cluster in advance through the above method, when the target cluster Ck matched by the current service is determined, the updated feature information Fi of the current service is input into the anomaly risk assessment model Mk corresponding to the target cluster Ck, thereby obtaining the anomaly probability of the current service. .

[0028] S103, determine the risk level adjustment coefficient of the current business based on the state cluster label, and determine the resource configuration parameters of the current business based on the anomaly probability and the risk level adjustment coefficient.

[0029] Optionally, the risk level adjustment coefficient for the current business is determined based on the status cluster label, including: determining the risk type of the current business based on the status cluster label, wherein the risk type includes low risk, medium risk and high risk; and determining the risk level adjustment coefficient for the current business based on the risk type.

[0030] Specifically, in this embodiment, for the current business, the risk type of the current business can be determined based on the status cluster label. The risk types include low risk, medium risk, and high risk. This embodiment pre-sets different risk level adjustment coefficients for different risk types. For example, the risk level adjustment coefficient set for high risk is... Risk level adjustment coefficient set for medium risk The risk level adjustment coefficient set for low risk is The values ​​decrease sequentially. This implementation does not limit the specific values ​​of the risk level adjustment coefficients corresponding to each risk type. Users can set them according to the actual configuration precision.

[0031] Optionally, the resource configuration parameters for the current business are determined based on the anomaly probability and risk level adjustment coefficient, including: obtaining the operating environment of the current business and determining the basic resource parameters and dynamic adjustment items based on the operating environment; obtaining the first product result of the anomaly probability and the first weight, and the second product result of the risk level adjustment coefficient and the second weight; and using the sum of the basic resource parameters, the dynamic adjustment items, the first product result, and the second product result as the resource configuration parameters for the current business.

[0032] Given that the probability of anomalies and the risk level adjustment coefficient for the current business are both determined, the resource configuration parameters for the current business can be calculated using the following formula (5): in, This refers to the identifier of the current business. This indicates the resource configuration parameters for the current business. Indicates basic resource parameters, Indicates a dynamically adjusted item. Indicates the probability of an anomaly. This represents the risk level adjustment factor. This represents the first weight assigned to the probability of an anomaly. This indicates the second weight set for the risk level adjustment coefficient. The aforementioned basic resource parameters and dynamic adjustment items are determined based on the current business operating environment. Since different businesses have different operating environments, the corresponding adjustments will vary. There may be differences for different services, but for the same batch of services, since the operating environment is roughly the same, the two parameters mentioned above are basically the same.

[0033] S104, Configure resources for the current service according to resource configuration parameters.

[0034] Specifically, in this embodiment, after obtaining the resource configuration parameters of the current service through the above method, resources can be configured for the current service according to the resource configuration parameters. The resource configuration parameters specifically include a specified number of resources to be configured, thus allowing a specified number of resources to be selected from the current system's resources and allocated to the current service. If the remaining resources in the system are greater than the specified number, then any specified number of resources can be arbitrarily selected from the remaining resources and allocated to the current service. However, if the remaining resources in the system are less than the specified number, it will be determined that the services currently allocated resources but in a dormant state will be identified. In this case, the resources occupied by the dormant services will be released, and the released resources will be allocated to the current service. Of course, this embodiment is merely an example and does not limit the specific method of resource configuration. As long as resource configuration can be achieved according to the resource configuration parameters, ensuring the normal execution of the current service, it is within the scope of protection of this application.

[0035] In this implementation, time-series features are used to cluster sample business data, thereby deeply mining the continuous dynamic information of the business and identifying deep-seated risk patterns and execution rules hidden in the time dimension. This achieves a significant shift in risk identification from static feature analysis to dynamic behavioral trajectory modeling, significantly improving the precision of clustering and business interpretability, and laying a solid foundation for building high-precision and robust prediction models. An improved sparrow algorithm is used to find the optimal hyperparameter combination. By combining long and short steps, the algorithm's ability to escape local optima is enhanced, helping to discover better hyperparameter configurations. It exhibits faster convergence speed and higher convergence while ensuring search quality. The system improves efficiency, and the acquired anomaly risk assessment model achieves enhanced prediction accuracy and generalization performance, enabling more accurate identification of businesses with different risk levels. It integrates the anomaly probability prediction results with risk levels based on state clustering to form a multi-dimensional, dynamically updated personalized resource allocation system. The system boasts a high degree of automation and intelligence, demonstrating strong engineering application potential. It achieves fully automated decision-making from data input, state clustering, model optimization, anomaly probability prediction to resource allocation parameter generation, reducing reliance on professional human experience. It transforms the traditional expert-experience-dependent coordination and grouping process into a standardized, automated process, supporting real-time decision-making needs and possessing strong industrial implementation capabilities.

[0036] The technical solution of this invention updates the features of the current business by matching the state cluster labels of the target clusters to enrich the feature information of the current business. Based on the updated feature information, a pre-trained anomaly risk assessment model is used to directly obtain the anomaly probability, thereby quickly identifying the operation status of the current business. Based on the anomaly probability and the risk level adjustment coefficient determined by the state cluster labels, the resource configuration parameters of the current business are comprehensively and accurately determined, so as to achieve accurate and efficient allocation of business resources according to the resource configuration parameters.

[0037] Example 2 Figure 2 This is a flowchart of another service resource configuration method provided by an embodiment of the present invention. Based on the above embodiment, after configuring resources for the current service according to resource configuration parameters, this embodiment further includes monitoring the execution process of the current service after resource configuration and evaluating the resource configuration result based on the execution status. Figure 2 As shown, the method includes: S201, based on the feature information of the current business, determine the matching target cluster from the cluster set, and concatenate the state cluster label of the target cluster with the feature information to obtain the updated feature information.

[0038] Optionally, before determining the matching target cluster from the cluster set based on the feature information of the current business, the method further includes: obtaining the feature information of the sample business, and clustering the sample business according to the feature information of the sample business to obtain multiple clusters; performing state recognition on each cluster to obtain the state of the cluster, and adding state cluster labels to the clusters according to the state; and constructing a cluster set based on the clusters with added state cluster labels.

[0039] Optionally, a matching target cluster can be determined from the cluster set based on the feature information of the current business, including: calculating the distance between the feature information and the center of each cluster in the cluster set; and taking the cluster with the smallest distance as the target cluster that matches the current business.

[0040] S202, update the feature information and input it into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability.

[0041] Optionally, before inputting the updated feature information into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability, the method further includes: obtaining a corresponding subset for each cluster, using an improved sparrow algorithm to obtain the optimal hyperparameter combination corresponding to the subset; configuring the gradient boosting decision tree framework according to the optimal hyperparameter combination, and using the subset to train the configured gradient boosting decision tree framework to obtain the anomaly risk assessment model corresponding to the cluster.

[0042] Optionally, an improved sparrow algorithm is used to obtain the optimal hyperparameter combination corresponding to the subset, including: randomly generating an initial population of a specified size for each subset; calculating the fitness of individuals and sorting them to determine the discoverer, the watcher, and the follower, as well as the random step size of the follower; updating the positions of the discoverer and the watcher, and updating the positions of the follower based on the random step size; evaluating the individuals after the position update, and taking the obtained global optimal solution as the optimal hyperparameter combination when the iteration condition is met.

[0043] S203, determine the risk level adjustment coefficient of the current business based on the status cluster label, and determine the resource configuration parameters of the current business based on the anomaly probability and the risk level adjustment coefficient.

[0044] Optionally, the risk level adjustment coefficient for the current business is determined based on the status cluster label, including: determining the risk type of the current business based on the status cluster label, wherein the risk type includes low risk, medium risk and high risk; and determining the risk level adjustment coefficient for the current business based on the risk type.

[0045] Optionally, the resource configuration parameters for the current business are determined based on the anomaly probability and risk level adjustment coefficient, including: obtaining the operating environment of the current business and determining the basic resource parameters and dynamic adjustment items based on the operating environment; obtaining the first product result of the anomaly probability and the first weight, and the second product result of the risk level adjustment coefficient and the second weight; and using the sum of the basic resource parameters, the dynamic adjustment items, the first product result, and the second product result as the resource configuration parameters for the current business.

[0046] S204, Configure resources for the current service based on resource configuration parameters.

[0047] S205 monitors the current business execution process after resource configuration and evaluates the resource configuration results based on the execution status.

[0048] Specifically, after configuring resources for the current service according to the resource configuration parameters, the current service will execute under the configured resources. This implementation monitors the execution process of the current service, specifically whether it executes normally. If the current service executes normally, it can be determined that the resources previously configured for it are correct. Conversely, if an anomaly occurs during execution, such as a resource shortage warning, it can be determined that the previously configured resources for the current service are insufficient. Therefore, this implementation can evaluate the previous resource configuration results by monitoring the execution of the current service, and can adjust and optimize the anomaly risk assessment model based on the evaluation results.

[0049] For example, if it is determined that there is insufficient resources during the current business execution, it can be determined that the output of the anomaly risk assessment model may be incorrect. In this case, a new sample business will be used to train the model. If the current business is executing normally, the current business will be used as a new sample business so that the anomaly risk assessment model can be adjusted according to the new sample business to ensure the real-time performance of the model and improve the prediction accuracy of the anomaly risk assessment model.

[0050] The technical solution of this invention updates the features of the current business by matching the state cluster labels of the target clusters to enrich the feature information of the current business. Based on the updated feature information, a pre-trained anomaly risk assessment model is used to directly obtain the anomaly probability, thereby quickly identifying the operation status of the current business. Based on the anomaly probability and the risk level adjustment coefficient determined by the state cluster labels, the resource configuration parameters of the current business are comprehensively and accurately determined, so as to achieve accurate and efficient allocation of business resources according to the resource configuration parameters.

[0051] Example 3 Figure 3 This is a schematic diagram of a resource allocation device for a service provided in an embodiment of the present invention. Figure 3 As shown, the device includes: an update feature information acquisition module 310, an anomaly probability assessment module 320, a resource configuration parameter determination module 330, and a resource configuration module 340.

[0052] Among them, the updated feature information acquisition module 310 is used to determine the matching target cluster from the cluster set based on the feature information of the current business, and to concatenate the state cluster label of the target cluster with the feature information to obtain the updated feature information; The anomaly probability assessment module 320 is used to input the updated feature information into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability; The resource configuration parameter determination module 330 is used to determine the risk level adjustment coefficient of the current business based on the status cluster label, and to determine the resource configuration parameters of the current business based on the anomaly probability and the risk level adjustment coefficient. The resource configuration module 340 is used to configure resources for the current business based on resource configuration parameters.

[0053] Optionally, the device also includes a clustering module for acquiring feature information of sample services and clustering sample services based on the feature information of sample services to obtain multiple clusters. State identification is performed on each cluster to obtain the state of the cluster, and state cluster labels are added to the clusters according to the state; Construct a cluster set based on the clusters with added state cluster labels.

[0054] Optionally, the updated feature information acquisition module is used to calculate the distance between the feature information and the center of each cluster in the cluster set; The cluster with the smallest distance is selected as the target cluster that matches the current business.

[0055] Optionally, the device also includes an anomaly risk assessment model training module, which is used to obtain the corresponding subset of data for each cluster and use an improved sparrow algorithm to obtain the optimal hyperparameter combination corresponding to the subset of data. The gradient boosting decision tree framework is configured based on the optimal hyperparameter combination, and the configured gradient boosting decision tree framework is trained using a subset of datasets to obtain the anomaly risk assessment model corresponding to the cluster.

[0056] Optionally, the anomaly risk assessment model training module is also used to randomly generate an initial population of a specified size for each subset of datasets; Calculate individual fitness and sort them to determine the discoverer, vigilant, and follower, as well as the random step size of the follower. Update the positions of the discoverer and vigilant, and update the positions of the follower based on the random step size. The updated individual is evaluated, and when the iteration conditions are met, the obtained global optimal solution is taken as the optimal hyperparameter combination.

[0057] Optionally, the resource configuration parameter determination module includes a risk level adjustment coefficient determination unit, which is used to determine the risk type of the current business based on the status cluster label, wherein the risk type includes low risk, medium risk and high risk; Determine the risk level adjustment factor for the current business based on the type of risk.

[0058] Optionally, the resource configuration parameter determination module includes a resource configuration parameter determination unit, which is used to obtain the current business operating environment and determine basic resource parameters and dynamic adjustment items based on the operating environment; Obtain the first product of the anomaly probability and the first weight, and the second product of the risk level adjustment coefficient and the second weight; The sum of the basic resource parameters, dynamic adjustment items, the first product result, and the second product result is used as the resource configuration parameters for the current business.

[0059] The resource configuration apparatus for a service provided in this embodiment of the invention can execute the resource configuration method for a service provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0060] Example 4 Figure 4A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. 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. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0061] 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.

[0062] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0063] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other electronic devices through computer networks such as the Internet and / or various telecommunications networks.

[0064] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as business resource allocation methods.

[0065] That is, based on the feature information of the current business, the matching target cluster is determined from the cluster set, and the status cluster label of the target cluster is concatenated with the feature information to obtain the updated feature information; The updated feature information is input into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability; The risk level adjustment coefficient for the current business is determined based on the state cluster label, and the resource configuration parameters for the current business are determined based on the anomaly probability and the risk level adjustment coefficient. Configure resources for the current service based on the resource configuration parameters.

[0066] In some embodiments, the resource allocation method for a service may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the resource allocation method for a service described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the resource allocation method for a service by any other suitable means (e.g., by means of firmware).

[0067] Various embodiments of the apparatuses and techniques described above herein can be implemented in digital electronic circuit devices, integrated circuit devices, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), device-on-a-chip (SoC) devices, complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable device including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage device, at least one input device, and at least one output device, and transmitting data and instructions to the storage device, the at least one input device, and the at least one output device.

[0068] Computer programs used to implement the resource allocation methods of this invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other non-stop data migration device, such that when executed by the processor, 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 a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0069] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution apparatus, device, or electronic device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media 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), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage electronics, magnetic storage electronics, or any suitable combination thereof.

[0070] To provide interaction with a user, the devices and techniques described herein can be implemented on an electronic device having: a display device (e.g., a touchscreen) for displaying information to the user; and buttons through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0071] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0072] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. 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 invention should be included within the scope of protection of this invention.

Claims

1. A resource allocation method for a business, characterized in that, The method includes: Based on the feature information of the current business, a matching target cluster is determined from the cluster set, and the state cluster label of the target cluster is concatenated with the feature information to obtain updated feature information; The updated feature information is input into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability; The risk level adjustment coefficient of the current service is determined based on the state cluster label, and the resource configuration parameters of the current service are determined based on the anomaly probability and the risk level adjustment coefficient. Configure resources for the current service according to the resource configuration parameters.

2. The method according to claim 1, characterized in that, Before determining the matching target cluster from the cluster set based on the feature information of the current business, the process also includes: Obtain feature information of sample services, and cluster the sample services according to the feature information to obtain multiple clusters; The state of each cluster is obtained by performing state identification on each cluster, and a state cluster label is added to the cluster according to the state; The cluster set is constructed based on the clusters with added state cluster labels.

3. The method according to claim 1, characterized in that, The step of determining the matching target cluster from the cluster set based on the feature information of the current business includes: Calculate the distance between the feature information and the center of each cluster in the cluster set; The cluster with the smallest distance is selected as the target cluster that matches the current service.

4. The method according to claim 1, characterized in that, Before inputting the updated feature information into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability, the method further includes: For each cluster, a corresponding subset of data is obtained, and an improved sparrow algorithm is used to obtain the optimal hyperparameter combination corresponding to the subset of data. The gradient boosting decision tree framework is configured according to the optimal hyperparameter combination, and the configured gradient boosting decision tree framework is trained using the subset dataset to obtain the anomaly risk assessment model corresponding to the cluster.

5. The method according to claim 4, characterized in that, The step of using the improved sparrow algorithm to obtain the optimal hyperparameter combination corresponding to the subset of data includes: For each of the aforementioned subsets of data, an initial population of a specified size is randomly generated; Calculate individual fitness and sort to determine the discoverer, the vigilant, and the follower, as well as the random step size of the follower; update the positions of the discoverer and the vigilant, and update the position of the follower based on the random step size; The individuals with updated positions are evaluated, and when the iteration conditions are met, the obtained global optimal solution is taken as the optimal hyperparameter combination.

6. The method according to claim 1, characterized in that, The step of determining the risk level adjustment coefficient for the current service based on the state cluster label includes: The risk type of the current service is determined based on the state cluster label, wherein the risk type includes low risk, medium risk, and high risk; The risk level adjustment factor for the current business is determined based on the risk type.

7. The method according to claim 1, characterized in that, The step of determining the resource configuration parameters for the current service based on the anomaly probability and the risk level adjustment coefficient includes: Obtain the current operating environment of the business, and determine the basic resource parameters and dynamic adjustment items based on the operating environment; Obtain the first product of the anomaly probability and the first weight, and the second product of the risk level adjustment coefficient and the second weight; The sum of the basic resource parameters, the dynamic adjustment item, the first product result, and the second product result is used as the resource configuration parameters for the current service.

8. A resource allocation device for a business, characterized in that, The device includes: The updated feature information acquisition module is used to determine the matching target cluster from the cluster set based on the feature information of the current business, and to concatenate the state cluster label of the target cluster with the feature information to obtain updated feature information; An anomaly probability assessment module is used to input the updated feature information into the anomaly risk assessment model corresponding to the target cluster to obtain the anomaly probability; The resource configuration parameter determination module is used to determine the risk level adjustment coefficient of the current service based on the status cluster label, and to determine the resource configuration parameters of the current service based on the anomaly probability and the risk level adjustment coefficient. The resource configuration module is used to configure resources for the current service according to the resource configuration parameters.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing 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 as described in any one of claims 1-7.

10. A storage medium for computer-executable instructions, wherein a computer program is stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

Images