Resource statistical system

Abstract

The invention discloses a resource statistical system, which comprises a resource statistical module used for counting the quantity and capacity of resources and evaluating the dynamic efficiency of the resources; the service publishing module is used for predicting service requirements and dynamically adjusting a publishing strategy in combination with the deep learning model; the service management module is used for realizing self-adaptive adjustment of the service by adopting a reinforcement learning algorithm; and the user behavior analysis module is used for grouping the users and carrying out personalized recommendation in combination with association rules.

Description

Technical Field

[0001] The present invention relates to the technical field of resource statistics, and in particular to a resource statistics system. Background Art

[0002] In recent years, with the rapid development of big data and artificial intelligence technology, resource management platforms are facing increasingly complex challenges. Traditional resource statistics and service management methods can no longer meet the current dynamic needs.

[0003] The existing technologies mainly have the following problems: The resource utilization efficiency evaluation lacks dynamism and cannot reflect the changes in resource usage in a timely manner. The data service release strategy is relatively fixed and difficult to adjust in real time according to user needs and system load. Service management lacks adaptive capabilities and cannot automatically optimize service configuration for different scenarios. User behavior analysis and personalized recommendation capabilities are insufficient, making it difficult to provide accurate service matching. To solve these problems, the industry has proposed some improvement plans, such as introducing machine learning algorithms for demand forecasting and using blockchain technology to improve data security. However, these plans often only address a single problem and lack systematicity and integrity. Summary of the invention

[0004] To achieve the above-mentioned purpose and other related purposes, the present invention discloses a resource statistics system, comprising: Resource statistics module, used to count resource quantity and capacity, and evaluate resource dynamic efficiency; The service publishing module is used to combine deep learning models to predict service demand and dynamically adjust the publishing strategy; The service management module is used to implement adaptive adjustment of services using reinforcement learning algorithms; The user behavior analysis module is used to group users and make personalized recommendations based on association rules.

[0005] Furthermore, the evaluation of resource dynamic efficiency includes: DEI(t) is the resource dynamic efficiency index, Ri(t) is the state of the i-th resource at time t, Fi(t) is the usage frequency of the resource at time t, US(t) is the user satisfaction at the current moment, and USmax is the maximum satisfaction.

[0006] Furthermore, predicting service demand and dynamically adjusting release strategy includes: SPF(s)=DNN(HSD,UBD,RSD)×RAC(s) Among them, SPF(s) is the service release optimization function, DNN is the deep neural network model, HSD is the historical service data, UBD is the user behavior data, RSD is the resource status data, RAC(s) is the real-time demand adjustment coefficient, and s is the service status.

[0007] Furthermore, the adaptive adjustment of the service includes: SAS(s)=RL(s,RF,EM)×SI(s) Among them, SAS(s) is the service adaptive adjustment strategy, RL is the reinforcement learning algorithm, RF is the reward function, EM is the environment model, SI(s) is the system stability indicator, and s is the service status.

[0008] Furthermore, the performing of personalized recommendation includes: UPR(u)=CA(u)×ARM(u,SD) Among them, UPR(u) is the user personalized recommendation for the u-th user, CA(u) is the clustering algorithm, ARM(u,SD) is the association rule mining, and SD is the service data.

[0009] By adopting the above technical solutions, this resource statistics system comprehensively utilizes historical data, real-time monitoring data and user behavior data to achieve all-round intelligent decision-making. The application of advanced algorithms such as deep learning, reinforcement learning, and cluster analysis has greatly improved the prediction accuracy and adaptability of the system. Through real-time efficiency evaluation, demand prediction and service adaptive adjustment, the system can quickly respond to environmental changes and maintain the optimal operating state. User satisfaction is taken as one of the core indicators to ensure that system optimization is directly reflected in the improvement of user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which:

[0011] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0013] Reference Figure 1The embodiment of the present invention provides a resource statistics system, including: The resource statistics module is used to count the number and capacity of resources and evaluate the dynamic efficiency of resources.

[0014] The resource statistics module is the foundation of the system and is responsible for real-time monitoring and evaluation of system resource usage. This module not only counts the number and capacity of resources, but more importantly, it achieves real-time optimization of resource usage through innovative dynamic efficiency evaluation methods.

[0015] This module calculates the resource dynamic efficiency index (DEI) using the following formula: Ri(t) represents the state of the i-th resource at time t, and possible values ​​include idle (0), partially occupied (0.5), and fully occupied (1). Fi(t) represents the frequency of use of the resource at time t, reflecting the importance of the resource. US(t) represents the user satisfaction at the current moment, which can be calculated through real-time user feedback or system performance indicators. USmax represents the preset maximum satisfaction value, which is used for normalization processing.

[0016] The innovation of this dynamic efficiency evaluation method lies in that it takes into account the dynamic changes of resource status and usage frequency. It introduces user satisfaction as a weight to directly link user experience with resource efficiency. Through normalization, the efficiency index at different time points is comparable.

[0017] The system can dynamically adjust the resource allocation strategy based on the real-time changes in DEI, such as triggering resource expansion when DEI is lower than a certain threshold, or considering resource recycling when DEI is continuously higher than a certain threshold, thereby optimizing resource utilization.

[0018] In this way, the system can monitor resource usage in real time and dynamically evaluate resource efficiency based on user satisfaction.

[0019] The service release module is used to combine deep learning models to predict service demand and dynamically adjust the release strategy.

[0020] The service publishing module combines deep learning models to predict service demand and dynamically adjust the publishing strategy. This module uses the following optimization functions:

[0021] SPF(s)=DNN(HSD,UBD,RSD)×RAC(s) SPF(s) is a service release optimization function, whose output determines the release priority and resource allocation of service s. DNN represents a deep neural network model, which can be a multi-layer perceptron (MLP), a long short-term memory network (LSTM) or a convolutional neural network (CNN). HSD is historical service data, including past service request volume, peak distribution, etc. UBD is user behavior data, such as user access patterns, residence time, etc. RSD is resource status data, which reflects the current system load. RAC(s) is a real-time demand adjustment coefficient, which is used to adjust the prediction results according to emergencies or temporary demands.

[0022] The workflow of this module is as follows: Data preprocessing: Clean, standardize and extract features from HSD, UBD and RSD. Model training: Use historical data to train the deep neural network model. Demand forecasting: Input the current UBD and RSD, and use the trained model to predict the service demand in the future. Strategy optimization: Combine RAC(s), calculate the final SPF(s) value, and determine the service release strategy.

[0023] The advantages of this method are: it can capture complex nonlinear relationships and improve prediction accuracy. By adjusting the coefficient RAC(s) in real time, it ensures the system's ability to respond quickly to emergencies. It takes into account user behavior, resource status and historical data to achieve multi-dimensional intelligent decision-making.

[0024] The deep neural network model predicts future service demand by analyzing historical data, user behavior and resource status. The real-time demand adjustment coefficient fine-tunes the prediction results according to the current situation, thereby achieving a more accurate service release strategy.

[0025] The service management module is used to adopt reinforcement learning algorithms to achieve adaptive adjustment of services.

[0026] The service management module uses reinforcement learning algorithms to achieve adaptive adjustment of services. The module uses the following strategies:

[0027] SAS(s)=RL(s,RF,EM)×SI(s) SAS(s) is the service adaptive adjustment strategy, which determines how to dynamically adjust the parameters and resource allocation of service s. RL represents the reinforcement learning algorithm, which can be Q-learning, SARSA or deep reinforcement learning algorithm such as DQN. RF is the reward function, which defines the optimization goal of the system, such as maximizing user satisfaction or minimizing resource consumption. EM is the environment model, which simulates the actual environment in which the service runs, including factors such as user requests and system load. SI(s) is the system stability indicator, which is used to balance optimization and stability.

[0028] The working principle of this module is as follows: State definition: Encode the current service state s as the input of the reinforcement learning algorithm. Action selection: Based on the current state, the RL algorithm selects an action (such as adding service instances, adjusting load balancing strategies, etc.). Action execution: The system executes the selected action and observes the changes in the environment. Reward calculation: The reward value of this action is calculated based on RF. Policy update: The RL algorithm updates its policy based on the reward obtained. Stability adjustment: Multiply the output of RL with SI(s) to ensure that the system does not lose stability due to over-optimization.

[0029] The innovation of this method is that the system can adapt to the complex and changing operating environment by continuously interacting with the environment. The system stability index is introduced to strike a balance between optimization and stability. The optimization goal of the system can be flexibly adjusted according to different reward function definitions. The reinforcement learning algorithm learns the optimal service adjustment strategy by continuously interacting with the environment. The system stability index ensures the overall stability of the system during the adjustment process.

[0030] The user behavior analysis module is used to group users and make personalized recommendations based on association rules.

[0031] The user behavior analysis module groups users and makes personalized recommendations based on association rules. This module uses the following methods:

[0032] UPR(u)=CA(u)×ARM(u,SD) UPR(u) is the personalized recommendation result for user u. CA(u) represents the clustering algorithm, which can be K-means, DBSCAN or hierarchical clustering. ARM(u, SD) represents the association rule mining algorithm, such as Apriori or FP-growth. SD is the service data, including the attributes of the service, the user's historical usage records, etc.

[0033] The workflow of this module is as follows: User feature extraction: Extract key features from the user's historical behavior and attribute information. User clustering: Use CA(u) to group users and identify user groups with similar behavior patterns. Association rule mining: For each user group, use ARM(u, SD) to mine the association rules of service usage. Personalized recommendation: Combine the characteristics of the user's group and the association rules to generate targeted service recommendations.

[0034] The advantages of this method are: through clustering algorithms, potential user groups can be discovered, improving the accuracy of recommendations. Association rule mining helps to discover non-obvious service usage patterns. Combining clustering and association rules not only takes into account the common characteristics of users, but also retains personalized features.

[0035] First, the system uses a clustering algorithm to group users, then analyzes the relationship between users and services through an association rule mining algorithm, and finally generates personalized service recommendations for each user.

[0036] Through the collaborative work of the above four modules, this resource statistics system comprehensively utilizes historical data, real-time monitoring data and user behavior data to achieve all-round intelligent decision-making. The application of advanced algorithms such as deep learning, reinforcement learning, and cluster analysis has greatly improved the system's prediction accuracy and adaptability. Through real-time efficiency evaluation, demand forecasting and service adaptive adjustment, the system can quickly respond to environmental changes and maintain the optimal operating state. It takes user satisfaction as one of the core indicators to ensure that system optimization is directly reflected in the improvement of user experience.

[0037] For the convenience of description, the above devices are described in terms of functions and are divided into various units and modules. Of course, when implementing the present application, the functions of each unit and module can be implemented in the same or multiple software and / or hardware.

[0038] Those skilled in the art will appreciate that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined.

[0039] For the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0040] It can be known from the description of the above implementation modes that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute the methods described in the various implementation modes of the present application or certain parts of the implementation modes.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A resource statistics system, characterized in that: include: Resource statistics module, used to count resource quantity and capacity, and evaluate resource dynamic efficiency; The service publishing module is used to combine deep learning models to predict service demand and dynamically adjust the publishing strategy; The service management module is used to implement adaptive adjustment of services using reinforcement learning algorithms; The user behavior analysis module is used to group users and make personalized recommendations based on association rules.

2. The system according to claim 1, characterized in that The evaluation of resource dynamic efficiency includes: ； DEI(t) is the resource dynamic efficiency index, Ri(t) is the state of the i-th resource at time t, Fi(t) is the usage frequency of the resource at time t, US(t) is the user satisfaction at the current moment, and USmax is the maximum satisfaction.

3. The system according to claim 1, characterized in that The predicting of service demand and dynamically adjusting the release strategy includes: SPF(s)=DNN(HSD,UBD,RSD)×RAC(s); Among them, SPF(s) is the service release optimization function, DNN is the deep neural network model, HSD is the historical service data, UBD is the user behavior data, RSD is the resource status data, RAC(s) is the real-time demand adjustment coefficient, and s is the service status.

4. The system according to claim 1, characterized in that The adaptive adjustment of the service includes: SAS(s)=RL(s,RF,EM)×SI(s); Among them, SAS(s) is the service adaptive adjustment strategy, RL is the reinforcement learning algorithm, RF is the reward function, EM is the environment model, SI(s) is the system stability indicator, and s is the service status.

5. The system according to claim 1, characterized in that The personalized recommendation comprises: UPR(u)=CA(u)×ARM(u,SD); Among them, UPR(u) is the user personalized recommendation for the u-th user, CA(u) is the clustering algorithm, ARM(u,SD) is the association rule mining, and SD is the service data.

Images