Resource allocation method and device, electronic equipment, storage medium and product
By training and fine-tuning the resource prediction model, and combining resource data from local and external data sources, the problem of accuracy in resource liquidity prediction was solved, a suitable resource allocation scheme was generated, and the efficiency and quality of resource allocation were improved.
Patent Information
- Application Number
- CN202511346682.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-19
- Publication Date
- 2026-01-02
AI Technical Summary
Existing technologies struggle to accurately predict resource liquidity, especially non-periodic and non-linear resource liquidity, leading to inappropriate resource allocation schemes and an inability to provide timely and accurate early warnings or adjustments, thus impacting user interests.
A trained and fine-tuned resource prediction model is used, which combines resource flow data from local data sources and resource allocation data from external data sources. A neural network model is constructed using machine learning or deep learning methods to predict resource inflow and outflow trends and generate appropriate resource allocation schemes.
It enables accurate prediction of resource liquidity, provides reliable resource allocation solutions, meets users' personalized needs, and improves the efficiency and quality of resource allocation.
Smart Images
Figure CN121256243A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and in particular to a resource allocation method, apparatus, electronic device, storage medium, and product. Background Technology
[0002] As living standards improve, people have increasingly diverse and demanding methods for managing and allocating resources. For a user, their resources can be used for various business purposes, such as storage in a resource management platform, resource exchange with other users, and conversion into different types of resources. Resources can be assets, hardware resources, software resources, network resources, or power resources, etc. For each business, the inflow and outflow of resources in the resource management platform are often dynamic rather than fixed, making them difficult to track. For example, users may periodically deposit a certain amount of resources into the platform and periodically use or consume a certain amount. By analyzing and predicting the inflow and outflow trends of resources, appropriate resource allocation decisions can be made for users. If the recommended resource allocation scheme is inappropriate, it may cause unnecessary losses for the user and make it difficult to provide timely and accurate early warnings or take appropriate prevention or control measures.
[0003] Currently, the Auto-regressive Moving Average (ARMA) model can be used to predict the flow of data objects in some scenarios. However, this model is suitable for periodically changing data and cannot capture non-periodic and non-linear characteristics. Furthermore, it can only use a single historical data point as input and cannot incorporate additional data, thus limiting its predictive power. Considering that resource flow is characterized by user subjectivity, non-periodicity, non-linearity, and susceptibility to external business data, accurately predicting resource flow and providing appropriate resource allocation solutions has become an urgent problem to be solved. Summary of the Invention
[0004] This application provides a resource allocation method, apparatus, electronic device, storage medium, and product to efficiently determine suitable resource allocation parameters and improve the quality and efficiency of resource allocation.
[0005] In a first aspect, embodiments of this application provide a resource allocation method, including:
[0006] The input data includes resource flow data within a set time period and resource allocation data for a specified business. The resource flow data is obtained from a local data source, and the resource allocation data is obtained from a non-local external data source through an application programming interface.
[0007] The input data is fed into a trained and fine-tuned resource prediction model to obtain resource prediction results for resource inflow and outflow. The resource prediction model is trained and fine-tuned based on sample data, which includes historical resource flow data obtained from local data sources and historical resource allocation data for the specified business obtained from non-local external data sources.
[0008] A resource allocation scheme is determined based on the resource prediction results. The resource allocation scheme includes at least one configuration type and the configuration quantity corresponding to each configuration type.
[0009] Secondly, embodiments of this application also provide a resource allocation device, including:
[0010] The acquisition module is used to acquire input data, which includes resource flow data within a set time period and resource allocation data for a specified business. The resource flow data is acquired from a local data source, and the resource allocation data is acquired from a non-local external data source through an application programming interface.
[0011] The prediction module is used to input the input data into a trained and fine-tuned resource prediction model to obtain resource prediction results for resource inflow and outflow. The resource prediction model is trained and fine-tuned based on sample data, which includes historical resource flow data obtained from local data sources and historical resource allocation data for the specified business obtained from non-local external data sources.
[0012] The configuration module is used to determine a resource configuration scheme based on the resource prediction results. The resource configuration scheme includes at least one configuration type and a configuration quantity corresponding to each configuration type.
[0013] Thirdly, embodiments of this application provide an electronic device, including:
[0014] One or more processors;
[0015] Storage device for storing one or more programs;
[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the resource allocation method as described in the first aspect.
[0017] Fourthly, embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the resource allocation method as described in the first aspect.
[0018] Fifthly, embodiments of this application also provide a computer program product, including a computer program and / or instructions, which, when executed by a processor, implement the resource configuration method as described in any of the above embodiments.
[0019] This application provides a resource allocation method, apparatus, electronic device, storage medium, and product. The resource allocation method includes: acquiring input data, which includes resource flow data within a set time period and resource allocation data for a specified business. The resource flow data is acquired from a local data source, and the resource allocation data is acquired from a non-local external data source via an application programming interface (API); inputting the input data into a trained and fine-tuned resource prediction model to obtain resource prediction results for resource inflows and outflows. The resource prediction model is trained and fine-tuned based on sample data, which includes historical resource flow data acquired from a local data source and historical resource allocation data for the specified business acquired from a non-local external data source; and determining a resource allocation scheme based on the resource prediction results. The resource allocation scheme includes at least one configuration type and a corresponding configuration quantity for each configuration type. The above technical solution utilizes a trained and fine-tuned high-precision resource prediction model, combined with resource flow data from local data sources and resource allocation data from external data sources for specific business operations. It comprehensively considers the impact of specific business operations and internal and external resource data on resource flow, and can accurately predict future resource inflow and outflow trends. This provides a reliable basis for determining resource allocation schemes, and on this basis, it can provide users with clear resource allocation schemes to meet their personalized resource allocation needs and enable them to allocate resources rationally. Attached Figure Description
[0020] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.
[0021] Figure 1 A flowchart illustrating a resource allocation method provided in an embodiment of this application;
[0022] Figure 2 A schematic diagram illustrating a process for constructing a resource prediction model, provided as an embodiment of this application;
[0023] Figure 3 This is a schematic diagram of the structure of a resource allocation device provided in an embodiment of this application;
[0024] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation
[0025] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present application, not the entire structure.
[0026] Before discussing the exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processes, many of these steps can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the steps can be rearranged. The process can be terminated when its operation is complete, but may also have additional steps not included in the figures. The process can correspond to a method, function, procedure, subroutine, subroutine, etc.
[0027] It should be noted that the concepts of "first" and "second" mentioned in the embodiments of this application are only used to distinguish different devices, modules, units or other objects, and are not used to limit the order of functions performed by these devices, modules, units or other objects or their interdependencies.
[0028] Furthermore, the embodiments and features described in this application may be combined with each other, unless otherwise specified.
[0029] The acquisition, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.
[0030] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the relevant content of the solution.
[0031] Figure 1 This is a flowchart illustrating a resource configuration method provided in an embodiment of this application. This embodiment is applicable to situations where resources need to be configured or allocated. Specifically, this resource configuration method can be executed by a resource configuration device, which can be implemented through software and / or hardware and integrated into an electronic device. The electronic device includes, but is not limited to, devices with data processing capabilities such as computers, smartphones, servers, or cloud platforms.
[0032] like Figure 1 As shown, the method specifically includes the following steps:
[0033] S110. Obtain input data, which includes resource flow data within a set time period and resource allocation data for a specified business. The resource flow data is obtained from a local data source, and the resource allocation data is obtained from a non-local external data source through an application programming interface.
[0034] In this embodiment, resource flow data can be understood as data reflecting the resource flow of a certain service within a recently defined time period, mainly including resource inflow data and resource outflow data. Resource flow data can be obtained from a local data source, which can be understood as a data source directly connected to the electronic device, or belonging to the same network or business system as the electronic device, such as a resource management platform. When obtaining resource flow data from a local data source, it can bypass external network interface forwarding or authorization verification. User resource configurations are recorded in the resource management platform, including when, where, and how much resource is stored in the resource management platform, or when and where how much resource is retrieved from the resource management platform. For example, for a service in the resource management platform (denoted as service A), resource flow data can include data on resource inflow and outflow for service A within a defined time period (such as the most recent year), specifically including the amount, frequency, and / or time of inflow or outflow of resources.
[0035] A designated service can be understood as a service associated with a service in the resource management platform (e.g., service A) but not managed by the resource management platform (e.g., service B), i.e., a service on the external network or in an external business system. Resources used for a designated service can be understood as resources extracted from services in the resource management platform and used for that service. Resource allocation data can be understood as data obtained from external data sources and used for resources in the designated service, such as the amount, frequency, and / or duration of resources extracted and used for the designated service. An external data source can be understood as a data source not directly connected to electronic devices, or belonging to a different network or business system than the electronic devices, such as a resource exchange or trading platform. Resource configuration data can be obtained by calling external networks or business systems through an Application Programming Interface (API). When obtaining resource flow data from external data sources, forwarding or authorization verification via an external network interface may be required.
[0036] The resource management platform can contain one or more business processes, and can also specify one or more business processes. For example, the resource management platform primarily records the resource inflow and outflow of business A. Some of the outflowing resources may be used for business B. It's important to note that this portion of the resource data is not stored in the resource management platform but needs to be obtained from business B's business platform. In other words, the resource allocation data for a specified business is external data to the resource management platform, and its data source differs from that of the resource flow data. In this embodiment, considering that external data can influence resource flow prediction, the resource allocation data for the specified business is also used as input data.
[0037] As an example, the business in the resource management platform can be the housing provident fund business, the resource can refer to the housing provident fund, the resource flow data can be the data on the deposit and use of the housing provident fund, the specified business can be the housing provident fund related transaction business, such as the housing transaction business, then the input data can include the data on the inflow and outflow of the housing provident fund, as well as the data on the housing provident fund used for the housing transaction business (such as the housing online contract registration data).
[0038] As an example, the business in the resource management platform can be the power resource business, where resources can refer to electrical energy, and resource flow data can be data on stored and used electrical energy. If the specified business can be the electrical energy trading business, then the input data can include data on stored and used electrical energy, as well as data on electrical energy used for the electrical energy trading business.
[0039] S120. Input the input data into a trained and fine-tuned resource prediction model to obtain resource prediction results for resource inflow and outflow. The resource prediction model is trained and fine-tuned based on sample data. The sample data includes historical resource flow data obtained from local data sources and historical resource allocation data for the specified business obtained from non-local external data sources.
[0040] For example, resource prediction models can be built based on machine learning or deep learning methods, such as neural network models or decision tree models. These models can be trained and fine-tuned based on sample data, which includes historical resource flow data obtained from local data sources and historical resource allocation data for a specific business obtained from external data sources. After training, the resource prediction model can learn the relationship between inputs (including resource flow data and resource allocation data for a specific business) and outputs (resource inflow or outflow trends, etc.). After fine-tuning, the model parameters are more suitable for the resource prediction scenario, possessing a deeper understanding and stronger reasoning ability regarding the characteristics and correlations of that scenario. Therefore, when given input resource flow data and resource allocation data for a specific business, it can accurately infer the resource inflow or outflow situation, thus obtaining reliable resource prediction results.
[0041] S130. Determine a resource allocation scheme based on the resource prediction results. The resource allocation scheme includes at least one configuration type and the configuration quantity corresponding to each configuration type.
[0042] In this context, configuration type can be understood as the method of configuring resources, such as periodic storage, flexible storage, retrieval and application to specific business operations, and / or lending. Configuration quantity can be understood as the amount of resources configured for the corresponding configuration type, such as periodically storing 1000 resources and / or applying 100 resources to specific business operations. Based on this, and according to the predicted resource inflow and outflow, a suitable and explicit resource configuration scheme can be provided to the user based on their historical resource configuration habits, patterns of resource inflow or outflow, and / or the user's requirements for remaining resources. For example, a user can store resources in a resource management platform to obtain certain benefits. Assuming the resource flow data is as follows: at the beginning of each month, users deposit 5,000 resources into the resource management platform; at the end of each month, users withdraw 1,000 resources from the resource management platform; and 2,000 resources are used for designated business purposes each month; the resource forecast result is: the resource management platform will have a surplus of 2,000 resources each month, and this surplus will remain largely unchanged over the next year; based on this, a resource allocation plan can be formulated, for example, storing 2,000 resources as periodic storage each month for the next year (i.e., they can only be withdrawn after the agreed period), so that users can obtain stable income. For example, resource flow data shows that users deposit 5,000 resources into the resource management platform each month, 2,000 of which are used for designated business purposes each month, and a large withdrawal of approximately 10,000 to 12,000 resources occurs every six months. Resource prediction results show that the resource management platform can have a surplus of 3,000 resources each month, and approximately 10,000 to 12,000 resources will be withdrawn every six months. Based on this, resource allocation scheme one can be formulated: deposit 500 resources as fixed-term deposits each month, or resource allocation scheme two: deposit 1,000 resources as current deposits each month (i.e., they can be withdrawn at any time), etc.
[0043] Understandably, when formulating resource allocation plans based on resource forecasting results, the goal can be to maximize the expected benefits for users. For example, for the resource allocation plan one and resource allocation plan two mentioned above, the benefits of the two resource allocation plans can be calculated by combining the rate of return of periodic storage and flexible storage, so as to select the optimal resource allocation plan.
[0044] It should be noted that, for ease of description, the resource flow data in the above examples has periodic or linear characteristics. In practical applications, for resource flow data with non-periodic or non-linear characteristics, the resource prediction model, as an artificial intelligence algorithm, has powerful prediction and reasoning capabilities and can also achieve accurate predictions.
[0045] This application provides a resource allocation method that utilizes a trained and fine-tuned high-precision resource prediction model. It combines resource flow data from local data sources with resource allocation data from external data sources for specific business operations. This comprehensively considers the impact of specific business operations and internal / external resource data on resource flow, accurately predicting future resource inflow and outflow trends. This provides a reliable basis for determining resource allocation schemes and, based on this, offers users clear resource allocation plans to meet their personalized resource allocation needs, enabling them to allocate resources rationally.
[0046] In one embodiment, the process of training and fine-tuning the resource prediction model includes:
[0047] S101. Obtain sample data and divide the sample data into training data and test data;
[0048] S102. Train the resource prediction model based on the training data;
[0049] S103. Test the trained resource prediction model based on the test data and obtain the test results;
[0050] S104. Based on the test results, fine-tune the trained resource prediction model to obtain the trained and fine-tuned resource prediction model.
[0051] In this embodiment, during the resource prediction model construction phase, sample data can be used to train and fine-tune the model. The sample data can include a large amount of historical resource flow data (obtained from local data sources) and historical resource allocation data for a specific business (obtained from external data sources). This sample data can be divided into two parts, for example, using 70% of the sample data as training data and 30% as test data. For the training process, historical resource flow data from the first time period (e.g., from five years ago to six months ago) and historical resource allocation data for the specified business can be used as input. An objective function or loss function is designed based on the difference between the resource flow situation inferred by the resource prediction model and the actual historical resource flow data from the second time period (e.g., the last six months). Through multiple iterative training iterations, the model parameters of the resource prediction model are adjusted to reduce losses and ensure prediction accuracy, ultimately resulting in the trained resource prediction model. For the fine-tuning process, the historical resource flow data of the first time period and the historical resource allocation data of the specified business in the test data can be used as inputs to test the difference between the resource flow situation inferred by the resource prediction model and the actual historical resource flow data of the second time period. The test results are then used to fine-tune the model parameters of the resource prediction model to further improve the performance of the resource prediction model and enable it to predict resource flow more accurately.
[0052] In one embodiment, the process of acquiring historical resource flow data includes:
[0053] S1. Read historical resource flow data from the resource management platform. The historical resource flow data includes first time series data corresponding to resource inflow and second time series data corresponding to resource outflow, which are statistically analyzed by time unit.
[0054] S2. Remove useless values, missing values, and / or outliers from the first and second time series data.
[0055] For example, historical resource flow data over a past period (such as the past five years) can be read from a resource management platform as sample data. This historical resource flow data is statistically analyzed by time unit, which can be year, month, day, week, or a set number of days. Time series data can include the amount of resources flowing in or out, along with corresponding timestamps. For instance, resources can refer to housing provident funds, the resource management platform can refer to the housing provident fund business system, and historical resource flow data can refer to monthly statistical time series data of housing provident fund inflows and outflows.
[0056] Data cleaning of the acquired historical resource flow data primarily aims to remove useless values (e.g., a point in time not falling within the selected time period), missing values (e.g., no information on resource inflow or outflow at a certain point in time), and / or outliers (e.g., obviously incorrect resource inflow or outflow amounts at a certain point in time). This improves the quality of the sample data, thereby enhancing the training performance of the resource prediction model, removing the influence of irrelevant or invalid information, and improving the inference performance of the resource prediction model.
[0057] In one embodiment, the process of acquiring historical resource flow data further includes:
[0058] S3. Convert the non-standard data in the first time series data and the second time series data into standard data; wherein, the non-standard data includes at least one of the following: data with irregular distribution, data with asymmetrical distribution, data with different dimensions, and data with heteroscedasticity.
[0059] For example, historical resource flow data may contain issues such as irregular original distributions of continuous variables, asymmetrical distributions, different dimensions among variables, and / or heteroscedasticity among variables. These problematic data can lead to decreased model fitting ability and affect the accuracy of model predictions. Data transformation (data standardization) of the acquired historical resource flow data aims to make the variable distribution more similar to a normal distribution. Discretization of continuous variables can be used to help improve the model's fitting effect. Specific transformation methods can include mathematical transformations, standardization and normalization, feature discretization, qualitative feature dummy coding, and / or time series data transformation. Based on this, the quality of the sample data can be further improved, enhancing the training effect and inference performance of the resource prediction model.
[0060] In one embodiment, the process of acquiring historical resource flow data further includes:
[0061] S4. Remove white noise data from the first time series data and the second time series data based on the Q statistic test method.
[0062] The Q statistic test, also known as the Box-Pierce Q test or Q test, is a statistical method used to test whether time series data contains white noise. Its core idea is to determine whether the data meets the characteristics of white noise by analyzing the autocorrelation of the time series data. For historical resource flow data, testing for and discarding white noise can further improve the quality of the sample data and enhance the training effect and inference performance of resource prediction models.
[0063] Figure 2 This is a schematic diagram illustrating a process for constructing a resource prediction model, as provided in one embodiment. Figure 2 As shown, the process of constructing a resource prediction model includes:
[0064] Step 1: Use the data interface to obtain historical resource flow data from the resource management platform, such as monthly data on the flow of various resources, including: historical resource inflow time series data and historical resource outflow time series data statistically analyzed on a monthly basis.
[0065] Step 2: Perform data cleaning on the above historical resource flow data, such as deleting useless values, missing values and / or outliers, or processing these values using fitting methods.
[0066] Step 3: For the above historical resource flow data, if one or two types of data have the following problems that may lead to a decrease in model fitting ability or affect prediction accuracy: such as irregular or asymmetrical distribution, or different dimensions between data, perform data transformation on the above types of time series data, such as standardizing and normalizing the data.
[0067] Step 4: Use the Q statistic test to determine white noise in the historical resource flow data. If the white noise hypothesis is met, the data is considered white noise, and training on the historical resource flow data is abandoned. If the white noise hypothesis is not met, the historical resource flow data is considered not white noise, and training on the historical resource flow data continues.
[0068] Step 5: Calculate the feature values of the above historical time series data, that is, encode the above historical time series data.
[0069] Step 6: Divide the above historical resource flow data into training data and test data.
[0070] Step 7: Use the encoded training data to model and train a machine learning algorithm (such as XgBoost) to generate a resource prediction model.
[0071] Step 8: Use the model trained in Step 7 to test the test dataset obtained in Step 6, calculate the Mean Absolute Percentage Error (MAPE) of the test results, and fine-tune the resource prediction model based on the test results.
[0072] Step 9: Using the resource inflow and outflow time series data (i.e., the first time series data and the second time series data) over a past period as input, input them into the resource prediction model obtained in Step 8, and output the resource prediction results for future resource flows.
[0073] In one embodiment, the resource prediction model is a decision tree model; the resource prediction model is trained based on the Extreme Gradient Boosting (XgBoost) algorithm; the corresponding objective function is determined based on the sum of the losses of each sample data and the regularization term of each leaf node in the decision tree, wherein the regularization term is determined based on the parameter learning rate, the number of leaf nodes and the vector formed by all leaf node values.
[0074] For example, the XgBoost algorithm is used for model training and prediction. During this process, historical resource flow data and external data (resource allocation data specific to the business) are used simultaneously. The external data influences the loss function portion of the objective function. The objective function can be:
[0075]
[0076] Where n is the number of samples, l is the loss per sample, and y i The true label value of the sample data. Ω(f) is the predicted value of the sample data. k ) is a regular expression term. γ and λ are the learning rate parameters, T is the number of leaf nodes, and w is the vector formed by the values of all leaf nodes in the decision tree. Predicted value Not only relying on historical resource flow data d cf It also relies on external data. hr Historical resource flow data d cf and external data d hr New matrices can be formed by stacking them horizontally, serving as independent variables for model training and prediction. The solution steps for the resource prediction model can refer to the conventional solution steps for the XgBoost model: such as Taylor second-order expansion, parameterizing the objective function, solving for the optimal leaf node weights, constructing the optimal tree, iterative optimization, etc., to finally obtain the trained resource prediction model.
[0077] Based on this, XGBoost is used to train the resource prediction model, which enables the resource prediction model to have better model performance and improve prediction accuracy.
[0078] In one embodiment, determining a resource allocation scheme based on resource prediction results includes: determining a resource allocation scheme based on resource prediction results and user-configured constraint information; wherein the constraint information includes at least one of the following: the proportion of available resources to required resources, and the amount of remaining resources; the resource allocation scheme includes at least one storage type and corresponding storage amount.
[0079] For example, constraint information can be understood as user requirements for resource configuration. For instance, it might require that the resource management platform have a minimum set amount of surplus resources each month (e.g., at least 1000 surplus resources for user use at any time), or that the resource configuration scheme must achieve a set amount of revenue (e.g., storing a certain amount of resources can generate revenue for the resource management platform). As another example, constraint information could be a requirement to achieve a certain ratio of available resources to required resources. Available resources can be understood as resources that exist and are readily available, while required resources can be understood as resources needed to ensure stable operation and support business operations. This ratio can serve as an indicator of whether the resource platform has a sufficient and stable source of resources, representing its ability to support the development of the corresponding business with long-term stable resources. The calculation formula is: Available stable resources / Stable resources required by the business. Taking the housing provident fund business as an example, this ratio can refer to the future net stable funding ratio. The future net stable funding ratio = (current month's end-of-month demand deposit balance + future month's fixed deposit maturity amount + future month's deposit amount + future month's loan principal and interest recovery) / (future month's withdrawal amount + future month's loan amount + loan amount to be disbursed + future month's other funds to be paid) × 100%.
[0080] The resource allocation scheme includes at least one configuration type and a corresponding configuration quantity. In this embodiment, the resource allocation method can be to obtain certain benefits through storage. Therefore, the configuration type can specifically be a storage type, such as one-month periodic storage, three-month periodic storage, six-month periodic storage and / or one-year periodic storage. Correspondingly, the configuration quantity can specifically be the storage quantity, such as the recommended storage quantity for one-month periodic storage, three-month periodic storage, six-month periodic storage and / or one-year periodic storage.
[0081] Based on this, personalized resource configuration schemes can be generated for users according to resource prediction results and user configuration constraints, to meet the needs of different users, such as ensuring the continuity or reliability of resource storage, so that users can obtain satisfactory benefits.
[0082] In one embodiment, a resource allocation scheme is determined based on resource prediction results and user-configured constraint information. Specifically, this includes: for different storage types and different storage volumes, an optimization algorithm is used to determine the resource allocation scheme based on the corresponding resource prediction results and constraint information; wherein, the optimization objective of the optimization algorithm is to maximize the expected revenue within a time unit.
[0083] In this embodiment, by combining resource flow data, resource prediction results, and constraint information, an optimization algorithm can be used to automatically solve for the optimal resource allocation scheme with the goal of maximizing the expected revenue within a time unit. Based on this, solving for the resource allocation scheme is equivalent to an optimization problem, which can be expressed as:
[0084] Maximize F(x,y / 3,z / 6,p / 12)
[0085] subject to x+y+z+p <dep c (t)
[0086] dep c (t+i)>dep cmin i = 1, ..., 12
[0087] NSFR(t+i)>NSFR min i = 1, ..., 12
[0088] Where x represents the amount of a one-month fixed deposit, y represents the amount of a three-month fixed deposit, z represents the amount of a six-month fixed deposit, and p represents the amount of a one-year fixed deposit; F(x,y / 3,z / 6,p / 12) is the expected interest earned on a one-month fixed deposit with amount x, a three-month fixed deposit with amount y, a six-month fixed deposit with amount z, and a one-year fixed deposit with amount p; depc (t) represents the remaining amount of resources currently stored; dep c (t+i) represents the predicted remaining resource quantity in the resource storage for the i-th month in the future; dep cmin This represents the minimum active storage limit set by the user in the previous step; NSFR(t+i) represents the predicted proportion of available resources to required resources in the i-th month of the future; NSFR min This indicates the lower limit of the proportion of future available resources to required resources set by the user in the previous step.
[0089] Based on this, by employing optimization algorithms to find the resource allocation scheme that maximizes benefits, users can gain significant benefits, meet their needs, and improve user satisfaction.
[0090] As an example, user-configured constraint information can be obtained through a visual interface. This constraint information includes the lower limit of the ratio of available resources to required resources. The visual interface provides an input box for "Future Net Stable Funding Ratio (Lower Limit)," allowing users to input the allowed lower limit of the future net stable funding ratio. The default value in the input box is 1.00, which users can modify or increase / decrease using the "+" and "-" buttons on either side of the input box. The single-step adjustment is 0.1 (the input value ranges from 0.01 to 100.00). Additionally, the future net stable funding ratio can also include the lower limit of remaining resources in demand deposits. This interface provides an input box for "Remaining Resources in Demand Deposits (Lower Limit)," allowing users to input the allowed lower limit of remaining resources in demand deposits (in ten thousand yuan). The default value in the input box is 1000, which users can modify or increase / decrease using the "+" and "-" buttons on either side of the input box. The single-step adjustment is 100 (the input value ranges from 1,000 to 1,000,000). The interface also provides a "Scheme Generation" button. After clicking this button, the system will automatically calculate and generate recommended storage amounts for "one-month fixed-term storage", "three-month fixed-term storage", "six-month fixed-term storage" and "one-year fixed-term storage", which will be displayed in the display box.
[0091] In one embodiment, the optimization algorithm used to determine the resource allocation scheme can be a grid search method.
[0092] Grid search is an exhaustive search method that searches for the optimal hyperparameters by traversing all possible combinations of hyperparameters. Using grid search to find the optimal resource allocation scheme offers high accuracy and efficiency, yielding a resource allocation scheme that maximizes user benefits and improves user satisfaction.
[0093] In one embodiment, the method further includes:
[0094] S140. Based on the resource forecast results, predict and display the changes in future liquidity indicators under the condition of changes in resource quantity for different types of resources.
[0095] In this embodiment, resource storage sensitivity analysis can be performed by combining resource forecasting results. That is, for periodic storage with different terms, the changes in future liquidity indicators (such as the ratio of available resources to required resources or the future net stable funding ratio) are calculated under the condition of changes in storage volume. Based on this, richer resource allocation functions can be provided, making it easier for users to fully grasp the future liquidity of resources.
[0096] As an example, a resource storage sensitivity analysis function can be provided through a visual interface. For instance, the interface could display the current current account storage amount; it could also provide a "Predict" button, which automatically predicts future resource liquidity indicators when clicked; and it could display a resource prediction list for the next 12 months (assuming the current month is T, then the predicted values for T+1 to T+12 months would be displayed), including "Remaining Current Account Resources at the End of the Month," "Term Deposit Resources," "Deposited Resources," "Withdrawal Resources," "Loan Principal and Interest Recovered," "Other Resources Required for Payment," and "Future Net Stable Funding Ratio." The interface could also provide time deposit... The storage quantity calculation input box allows users to enter the resource quantity for "one-month fixed-term storage", "three-month fixed-term storage", "six-month fixed-term storage", and "one-year fixed-term storage" (unit: RMB 10,000; default value is 100, users can enter and modify it, or increase or decrease it using the "+" and "-" buttons on both sides of the input box, with a single step adjustment of 10, and the modifiable value range is 1 to 1,000,000). The interface also provides a "Calculate" button. After clicking this button, users can perform the following operations: update the predicted values of relevant parameters; calculate and update the predicted value of "future net stable funding ratio" for the next 12 months (including the current month).
[0097] In one embodiment, the method further includes:
[0098] S150. Register and display the rate of return and corresponding activation date of different types of resources through the first resource ledger;
[0099] S160. Add, edit, or delete yield rates and corresponding activation dates based on user operations.
[0100] In this embodiment, the first resource ledger can be used to register different types (such as different periods) of yield and the corresponding activation date, and display the resource storage yield in a list. In addition, it also supports user operations such as adding, editing or deleting.
[0101] For example, in the visualization interface of the first resource ledger, users can add yield rate records by clicking the "Add" button, select the storage type (one-month fixed-term storage, three-month fixed-term storage, six-month fixed-term storage, one-year fixed-term storage, two-year fixed-term storage, three-year fixed-term storage, five-year fixed-term storage) from the drop-down menu, and enter the yield rate value. After adding the record, a row is automatically added to the list, displaying the activation date, which defaults to the date (day T) on which the "Create" action was performed. Users can also modify the interest rate value and activation date of the yield rate record by clicking the "Edit" button in the operation bar. Users can also delete the selected yield rate record by clicking the "Delete" button in the operation bar. After clicking the "Delete" button, the system will display a prompt box asking whether to confirm the deletion.
[0102] Based on this, by registering and displaying the rate of return, the resource allocation process becomes more functional and the resource information is more intuitive, making it easier for users to understand the rate of return and adjust the resource allocation plan, and supporting convenient user operations.
[0103] In one embodiment, the method further includes:
[0104] S170. Register and display resource storage information through the second resource ledger;
[0105] S180. Add, edit, or delete resource storage information based on user operations;
[0106] The resource storage information includes storage type, storage quantity, yield rate, deposit date, and expiration date, with the expiration date determined based on the storage type and deposit date.
[0107] In this embodiment, the second resource ledger can be used to register resource storage status and is displayed in the form of a list. In addition, it also supports user operations such as adding, editing or deleting.
[0108] For example, in the visualization interface of the second resource ledger, users can add storage records by clicking the "Add" button, selecting the category (one-month fixed-term deposit, three-month fixed-term deposit, six-month fixed-term deposit, one-year fixed-term deposit, two-year fixed-term deposit, three-year fixed-term deposit, five-year fixed-term deposit) from the drop-down list, and filling in the resource amount, interest rate, and deposit date. After adding the record, a row is automatically added to the list, displaying the relevant information, where the "Maturity / Withdrawal Date" is automatically calculated based on the storage type and deposit date. Users can also modify the relevant information of the storage record by clicking the "Edit" button in the operation bar. If there is an early withdrawal of a fixed-term deposit (i.e., the withdrawal date is earlier than the maturity date), the "Maturity / Withdrawal Date" field can be edited, and the withdrawal date can be filled in. Users can also delete the selected deposit record by clicking the "Delete" button in the operation bar. After clicking the "Delete" button, the system will display a prompt box asking whether to confirm the deletion.
[0109] Based on this, by registering and displaying resource storage status, the resource allocation process becomes more feature-rich, resource information becomes more intuitive, making it easier for users to grasp the rate of return and adjust resource allocation plans, and also supporting convenient user operations.
[0110] The resource prediction and allocation method of this application embodiment provides the following functions: five parts: resource prediction model prediction, first resource ledger, second resource ledger, resource storage sensitivity analysis, and resource allocation scheme generation. The resource prediction model is used to predict future resource inflows and outflows; the first resource ledger is used to register the interest rates of different types of storage; the second resource ledger is used to register each periodic storage transaction in the center; the resource storage sensitivity analysis is used to perform sensitivity analysis on resource storage schemes; the resource allocation scheme generation function can calculate and automatically generate the optimal storage recommendation scheme based on the resource prediction value, the rate of return, and the second resource ledger, while meeting the resource liquidity requirements.
[0111] The resource prediction and allocation method of this application can achieve resource prediction (such as predicting monthly data on resource inflows or outflows from the housing provident fund center); introduce external data (such as housing online contract registration data), capture the correlation between external data and resource flow, and predict resource flow based on the impact of external data, which can improve the prediction accuracy of the resource prediction model; perform resource storage sensitivity analysis, allowing users to manually calculate the impact of different types and amounts of storage on resource liquidity indicators; and automatically generate the optimal resource storage scheme under the condition of resource liquidity, with rich functions, and can meet user needs and improve user satisfaction.
[0112] Figure 3 This is a schematic diagram of a resource allocation device provided in an embodiment of this application. Figure 3As shown, the resource configuration device provided in this embodiment includes:
[0113] The acquisition module 210 is used to acquire input data, which includes resource flow data within a set time period and resource allocation data for a specified business. The resource flow data is acquired from a local data source, and the resource allocation data is acquired from a non-local external data source through an application programming interface.
[0114] Prediction module 220 is used to input the input data into the resource prediction model to obtain resource prediction results for resource inflow and outflow;
[0115] Configuration module 230 is used to determine a resource configuration scheme based on the resource prediction results.
[0116] This device utilizes a trained and fine-tuned high-precision resource prediction model, combined with resource flow data from local data sources and resource allocation data from external data sources for specific services. It comprehensively considers the impact of specific services and internal and external resource data on resource flow, and can accurately predict future resource inflow and outflow trends. This provides a reliable basis for determining resource allocation schemes, and on this basis, it can provide users with clear resource allocation schemes to meet users' personalized resource allocation needs and enable users to allocate resources rationally.
[0117] Based on any of the above embodiments, the device further includes: a training fine-tuning module, which specifically includes:
[0118] An acquisition unit is used to acquire sample data, which is divided into training data and test data.
[0119] Training unit, used to train resource prediction model based on the training data;
[0120] The testing unit is used to test the trained resource prediction model based on the test data and obtain test results.
[0121] The fine-tuning unit is used to fine-tune the trained resource prediction model based on the test results, so as to obtain a trained and fine-tuned resource prediction model.
[0122] Based on any of the above embodiments, the acquisition unit includes:
[0123] The reading subunit is used to read historical resource flow data from the resource management platform. The historical resource flow data includes a first time series data corresponding to resource inflow and a second time series data corresponding to resource outflow, which are statistically analyzed by time unit.
[0124] The cleaning subunit is used to remove useless values, missing values, and / or outliers from the first time series data and the second time series data.
[0125] Based on any of the above embodiments, the acquisition unit further includes:
[0126] A conversion subunit is used to convert non-standard data in the first time series data and the second time series data into standard data;
[0127] The non-standard data includes at least one of the following: data with irregular distribution, data with asymmetrical distribution, data with different dimensions, and data with heteroscedasticity.
[0128] Based on any of the above embodiments, the acquisition unit further includes:
[0129] The sub-unit removal method is used to remove data that is white noise from the first time series data and the second time series data based on the Q statistic test method.
[0130] Based on any of the above embodiments, the resource prediction model is a decision tree model; the resource prediction model is trained based on the extreme gradient boosting algorithm; the corresponding objective function is determined according to the sum of the losses of each sample data and the regularization term of each leaf node in the decision tree, and the regularization term is determined according to the parameter learning rate, the number of leaf nodes and the vector formed by all leaf node values.
[0131] Based on any of the above embodiments, the configuration module 230 is specifically used to: determine a resource configuration scheme according to the resource prediction results and the constraint information configured by the user; the constraint information includes at least one of the following: the proportion of available resources to required resources, and the amount of remaining resources.
[0132] Based on any of the above embodiments, the configuration module 230 is specifically used for:
[0133] For different storage types and storage volumes, an optimization algorithm is used to determine the resource allocation scheme based on the corresponding resource prediction results and constraint information.
[0134] The optimization objective of the optimization algorithm is to maximize the expected revenue within a unit of time.
[0135] Based on any of the above embodiments, the optimization algorithm includes a grid search method.
[0136] Based on any of the above embodiments, the device further includes:
[0137] The sensitivity prediction module is used to predict and display the changes in future liquidity indicators under changes in resource quantity, based on the resource prediction results, for different types of resources.
[0138] Based on any of the above embodiments, the device further includes:
[0139] The first registration module is used to register and display the rate of return and corresponding activation date of different types of resources through the first resource ledger;
[0140] The first operation module is used to add, edit, or delete yield rates and corresponding activation dates based on user operations.
[0141] Based on any of the above embodiments, the device further includes:
[0142] The second registration module is used to register and display resource storage information through the second resource ledger;
[0143] The second operation module is used to add, edit, or delete resource storage information based on user operations;
[0144] The resource storage information includes storage type, storage quantity, yield rate, deposit date, and expiration date, with the expiration date determined based on the storage type and deposit date.
[0145] The resource configuration apparatus provided in this application embodiment can be used to execute the resource configuration method provided in any of the above embodiments, and has corresponding functions and beneficial effects.
[0146] Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of this application, is shown. The electronic device 10 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 10 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, user equipment, 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 application described and / or claimed herein.
[0147] like Figure 4As 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.
[0148] 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 devices through computer networks such as the Internet and / or various telecommunications networks and wireless networks.
[0149] 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.
[0150] In some embodiments, the methods described above can 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 can 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 methods described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the methods of any of the embodiments described above by any other suitable means (e.g., by means of firmware).
[0151] Various embodiments of the systems and techniques described above herein 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), payload-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 system 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 system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.
[0152] Computer programs used to implement the methods of this application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0153] In the context of this application, 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 system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can 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 devices, magnetic storage devices, or any suitable combination of the foregoing.
[0154] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device 10, which includes: 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 pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device 10. 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).
[0155] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.
[0156] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the 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 cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.
[0157] This application also provides a computer program product, including a computer program and / or instructions, which, when executed by a processor, implement the resource configuration method as described in any of the above embodiments.
[0158] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this application can be achieved, and this is not limited herein.
[0159] The specific embodiments described above do not constitute a limitation on the scope of protection 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 should be included within the scope of protection of this application.
Claims
1. A resource allocation method, characterized in that, include: The input data includes resource flow data within a set time period and resource allocation data for a specified business. The resource flow data is obtained from a local data source, and the resource allocation data is obtained from a non-local external data source through an application programming interface. The input data is fed into a trained and fine-tuned resource prediction model to obtain resource prediction results for resource inflow and outflow. The resource prediction model is trained and fine-tuned based on sample data, which includes historical resource flow data obtained from local data sources and historical resource allocation data for the specified business obtained from non-local external data sources. A resource allocation scheme is determined based on the resource prediction results. The resource allocation scheme includes at least one configuration type and the configuration quantity corresponding to each configuration type.
2. The method according to claim 1, characterized in that, The training and fine-tuning process of the resource prediction model includes: Acquire sample data and divide the sample data into training data and test data; Train the resource prediction model based on the training data; The trained resource prediction model is tested based on the test data to obtain test results; Based on the test results, the trained resource prediction model is fine-tuned to obtain a trained and fine-tuned resource prediction model.
3. The method according to claim 2, characterized in that, The process of acquiring historical resource flow data includes: Historical resource flow data is read from the resource management platform. The historical resource flow data includes a first time series data corresponding to resource inflow and a second time series data corresponding to resource outflow, which are statistically analyzed by time unit. Remove useless values, missing values, and / or outliers from the first time series data and the second time series data.
4. The method according to claim 3, characterized in that, The process of acquiring historical resource flow data also includes: Convert the non-standard data in the first time series data and the second time series data into standard data; The non-standard data includes at least one of the following: data with irregular distribution, data with asymmetrical distribution, data with different dimensions, and data with heteroscedasticity.
5. The method according to claim 3, characterized in that, The process of acquiring historical resource flow data also includes: The Q statistic test method was used to remove white noise data from the first and second time series data.
6. The method according to claim 2, characterized in that, The resource prediction model is a decision tree model; the resource prediction model is trained based on the extreme gradient boosting algorithm; the corresponding objective function is determined based on the sum of the losses of each sample data and the regularization term of each leaf node in the decision tree, and the regularization term is determined based on the parameter learning rate, the number of leaf nodes and the vector formed by all leaf node values.
7. The method according to claim 1, characterized in that, The step of determining the resource allocation scheme based on the resource prediction results includes: Based on the resource prediction results and the user-configured constraint information, a resource allocation scheme is determined; The constraint information includes at least one of the following: the proportion of available resources to required resources, and the amount of remaining resources; The resource allocation scheme includes at least one storage type and corresponding storage capacity.
8. The method according to claim 7, characterized in that, The step of determining a resource allocation scheme based on the resource prediction results and user-configured constraint information includes: For different storage types and storage volumes, an optimization algorithm is used to determine the resource allocation scheme based on the corresponding resource prediction results and constraint information. The optimization objective of the optimization algorithm is to maximize the expected revenue within a unit of time.
9. The method according to claim 8, characterized in that, The optimization algorithm includes grid search.
10. The method according to claim 1, characterized in that, Also includes: Based on the resource forecast results, for different types of resources, the changes in future liquidity indicators under changes in resource quantity are predicted and displayed.
11. The method according to claim 1, characterized in that, Also includes: The first resource ledger registers and displays the rate of return and corresponding activation date of different types of resources; Users can add, edit, or delete yield rates and corresponding activation dates based on their actions.
12. The method according to claim 1, characterized in that, Also includes: Resource storage information is registered and displayed through a second resource ledger; Add, edit, or delete resource storage information based on user actions; The resource storage information includes storage type, storage quantity, yield rate, deposit date, and expiration date, with the expiration date determined based on the storage type and deposit date.
13. A resource allocation device, characterized in that, include: The acquisition module is used to acquire input data, which includes resource flow data within a set time period and resource allocation data for a specified business. The resource flow data is acquired from a local data source, and the resource allocation data is acquired from a non-local external data source through an application programming interface. The prediction module is used to input the input data into a trained and fine-tuned resource prediction model to obtain resource prediction results for resource inflow and outflow. The resource prediction model is trained and fine-tuned based on sample data, which includes historical resource flow data obtained from local data sources and historical resource allocation data for the specified business obtained from non-local external data sources. The configuration module is used to determine a resource configuration scheme based on the resource prediction results. The resource configuration scheme includes at least one configuration type and a configuration quantity corresponding to each configuration type.
14. An electronic device, characterized in that, include: At least one processor; A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the resource allocation method as described in any one of claims 1-12.
15. 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 resource configuration method as described in any one of claims 1-12.
16. A computer program product comprising a computer program and / or instructions, characterized in that, When the computer program and / or instructions are executed by the processor, they implement the resource allocation method as described in any one of claims 1-12.