Intelligent dynamic adjustment method for flow data

Abstract

The invention relates to the technical field of data traffic control, and discloses an intelligent dynamic adjustment method for traffic data, and the method sequentially comprises the steps: collecting traffic use data from each data source; preprocessing the collected traffic use data, and extracting key indexes for subsequent analysis; performing prediction analysis on the preprocessed data by using the trained machine learning model to obtain a traffic demand prediction value in a period of time in the future; calculating an optimal flow distribution scheme by using an optimization algorithm according to a flow prediction result and a current flow pool resource condition; automatically adjusting resource allocation of the traffic pool according to the formulated adjustment strategy, wherein the resource allocation comprises increasing or reducing traffic quotas of the users; and the traffic use condition is continuously monitored, an alarm mechanism is immediately triggered when an abnormal condition is found, and related personnel are notified to handle the abnormal condition. By adopting the intelligent dynamic adjustment method for the traffic data, the traffic use result prediction efficiency can be improved, the traffic distribution effect can be optimized, and the user stickiness can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of data flow control, and in particular to an intelligent dynamic adjustment method for flow data. Background Art

[0002] In the current information age, with the rapid development of mobile Internet, traffic cards, as an important bridge connecting users and the Internet, have become one of the core businesses of major telecom operators and network service providers in terms of sales, operation, monitoring and management. There are already a variety of traffic pool operation service systems on the market, which mainly implement the basic management functions of traffic cards, such as inventory management, sales records, user activation, etc. However, these systems generally have the following deficiencies in real-time monitoring and dynamic adjustment:

[0003] 1. Insufficient real-time monitoring capabilities: Traditional systems often rely on scheduled tasks or user-initiated triggers to update the traffic pool status, and are unable to reflect the dynamic changes in traffic usage in real time, resulting in delayed system response and reduced user experience during peak hours or sudden traffic demands.

[0004] 2. Inflexible resource allocation: Most existing systems use static traffic allocation strategies, which means that the capacity and allocation rules of each traffic pool are pre-set, and lack the ability to dynamically adjust according to real-time needs. This approach easily leads to uneven resource allocation, with some traffic pool resources being idle while others are insufficient, affecting user experience.

[0005] 3. Insufficient prediction and decision support: Traditional systems lack intelligent prediction modules and cannot accurately predict the traffic demand trend in the future. Therefore, when formulating traffic allocation strategies, they often rely on experience and judgment, which lacks scientific basis. At the same time, the system also lacks decision support functions based on real-time data, making it difficult to adjust strategies in time to deal with emergencies.

[0006] 4. High operation and maintenance costs: Due to insufficient real-time monitoring capabilities and inflexible resource allocation, traditional systems often require a lot of manpower and material resources for operation and maintenance management, including regular manual adjustment of traffic pool configurations, handling user complaints, etc., which increases the company's operating costs. Summary of the invention

[0007] The purpose of this application is to provide an intelligent dynamic adjustment method for flow data to solve the technical problems raised in the above background.

[0008] To achieve the above purpose, the present application discloses the following technical solution: a method for intelligent dynamic adjustment of flow data, the method comprising the following steps in sequence:

[0009] Data collection: Collect traffic usage data from various data sources;

[0010] Data processing: pre-process the collected traffic usage data and extract key indicators for subsequent analysis;

[0011] Traffic forecasting: Use the trained machine learning model to perform forecast analysis on the preprocessed data to obtain the traffic demand forecast for a period of time in the future;

[0012] Dynamic adjustment strategy formulation: Based on the traffic prediction results and the current traffic pool resource status, the optimization algorithm is used to calculate the optimal traffic allocation plan;

[0013] Execute adjustment: Automatically adjust the traffic pool resource allocation according to the established adjustment strategy, including increasing or decreasing the user's traffic quota;

[0014] Real-time monitoring and alarm: Continuously monitor traffic usage, immediately trigger the alarm mechanism when abnormal situations are found, and notify relevant personnel to handle them.

[0015] Preferably, in the data collection, the traffic usage data is collected from each data source periodically or according to trigger conditions through distributed data collection nodes.

[0016] Preferably, in the data processing, the preprocessing includes cleaning, deduplication and integration.

[0017] Preferably, in the traffic prediction, the training step of the machine learning model is: using a machine learning algorithm to build a traffic prediction model, training historical traffic data, and generating a machine learning model for predicting future traffic demand.

[0018] Preferably, when training the machine learning model, time, user behavior patterns, network conditions and corresponding historical traffic data are also used as model training data.

[0019] Preferably, the optimization algorithm is a genetic algorithm or a simulated annealing algorithm.

[0020] Preferably, in the real-time monitoring and alarm, abnormality detection is performed on traffic usage according to set abnormality detection rules, and the abnormality includes limiting traffic or suspending service.

[0021] Preferably, the intelligent dynamic adjustment method for flow data further includes:

[0022] Traffic usage, user behavior, and operating performance are displayed through monitoring views.

[0023] Compared with the prior art, the intelligent dynamic adjustment method of flow data in this application has the following technical effects:

[0024] 1. Accurate prediction and efficient adjustment: By integrating advanced machine learning models, the intelligent dynamic adjustment algorithm can accurately predict future traffic demand, and the prediction accuracy has been effectively improved. At the same time, the dynamic adjustment strategy based on the prediction results can quickly respond to market changes, reduce resource waste, and improve resource utilization.

[0025] 2. Automation and intelligence: The algorithm realizes the automation and intelligence of the entire process from data collection, processing to prediction and adjustment, which greatly reduces the need for manual intervention, reduces human errors, and improves the stability and reliability of the system. In addition, the self-learning ability of the algorithm enables it to continuously optimize and adjust strategies to adapt to the complex and changing market environment.

[0026] 3. Multi-dimensional analysis and personalized services: The introduction of multi-dimensional data analysis functions enables the system to more comprehensively understand user needs and market dynamics, and provide users with personalized traffic management services. This customized service not only improves user satisfaction, but also enhances user stickiness, bringing more business opportunities to enterprises.

[0027] 4. Real-time monitoring and rapid response: The real-time monitoring module can promptly detect and handle abnormal situations in traffic usage to ensure the legality and rationality of traffic usage. The rapid response mechanism further shortens the problem handling time and reduces potential economic losses and social impacts. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0029] Figure 1 A flow chart of a method for intelligent dynamic adjustment of flow data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present application are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0031] In this article, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of more restrictions, the elements defined by the sentence "comprising..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0032] This embodiment provides an intelligent dynamic adjustment method for traffic data, which integrates efficient data processing, accurate prediction and intelligent decision-making functions, aiming to achieve refined management and efficient use of traffic resources. Figure 1 As shown, the method comprises the following steps in sequence:

[0033] Data collection: Collect traffic usage data from various data sources;

[0034] Data processing: pre-process the collected traffic usage data and extract key indicators for subsequent analysis;

[0035] Traffic forecasting: Use the trained machine learning model to perform forecast analysis on the preprocessed data to obtain the traffic demand forecast for a period of time in the future;

[0036] Dynamic adjustment strategy formulation: Based on the traffic prediction results and the current traffic pool resource status, the optimization algorithm is used to calculate the optimal traffic allocation plan;

[0037] Execute adjustment: Automatically adjust the traffic pool resource allocation according to the established adjustment strategy, including increasing or decreasing the user's traffic quota;

[0038] Real-time monitoring and alarm: Continuously monitor traffic usage, immediately trigger the alarm mechanism when abnormal situations are found, and notify relevant personnel to handle them.

[0039] By using the above method, the corresponding system structure is constructed. The system integrates modules such as intelligent dynamic adjustment algorithm, real-time data collection and processing, traffic prediction, threshold setting, resource optimization allocation, and real-time alarm and intervention mechanism into a complete traffic pool operation service system. Among them, the intelligent dynamic adjustment method of traffic data, that is, an intelligent dynamic adjustment algorithm, is the core of the system, which integrates three functional modules: traffic prediction, threshold setting, and resource optimization allocation. The algorithm realizes intelligent management of traffic resources by continuously monitoring traffic usage and combining multi-dimensional data such as user behavior patterns and network conditions.

[0040] In this embodiment, the system collects traffic usage data in real time through distributed data collection nodes, and uses data cleaning and preprocessing technology to ensure the accuracy and integrity of the data. The preprocessed data is pushed to the central processing system to provide a basis for subsequent analysis and prediction. It is feasible that the preprocessing of the data includes but is not limited to: abnormal traffic detection and time series decomposition.

[0041] Among them, the content of abnormal traffic detection is: identify and filter out abnormal values ​​in network traffic through anomaly detection algorithms (such as Isolation Forest, One-Class SVM, etc.). Outliers may be caused by network attacks, equipment failures, etc. If they are not handled, they will affect the training effect of the model. In the prior art, most existing models directly use raw data for training, ignoring the impact of outliers on model performance. This method improves the robustness and accuracy of the model by preprocessing the data.

[0042] In addition, the content of time series decomposition is: using time series decomposition technology (such as STL Decomposition) to decompose network traffic data into three parts: trend, seasonality and noise. For each part, a different LSTM model is used for modeling. In the prior art, traditional LSTM models often directly model the original time series data, and it is difficult to effectively separate the different components in the data. This method uses decomposition technology to allow each LSTM model to focus on processing a specific type of pattern, thereby improving the overall prediction accuracy.

[0043] In traffic forecasting, machine learning algorithms (such as LSTM or ARIMA) are used to build traffic forecasting models, train historical traffic data, and generate models that can predict future traffic demand. The forecasting model considers multiple factors such as time, user behavior patterns, and network conditions to improve the accuracy of the forecast. It is feasible that a partial computer program using the LSTM model for traffic forecasting is as follows:

[0044]

[0045] In order to further ensure the accuracy of the prediction results, the scheme of this embodiment also includes building a multi-task learning framework, in which one task is to predict the network traffic in the future, and the other task is to predict the trend of network traffic. The two tasks share the underlying LSTM layer, but have their own independent output layers. In this way, the use of multi-task learning can help the model learn richer feature representations while improving the performance of the main task (i.e., network traffic prediction). This method can make full use of the information in the data better than optimizing a single task alone. In addition, the attention mechanism is introduced in the LSTM model, allowing the model to dynamically adjust the weights of past time steps according to the importance of the current time step. Secondly, the Shapley value or other interpretability tools can be used to analyze the model decision process, provide users with clear prediction basis instructions, and increase the transparency and credibility of the model.

[0046] When formulating dynamic adjustment strategies, the system automatically sets and adjusts the traffic usage threshold based on traffic forecast results. The threshold setting takes into account factors such as user priority and demand urgency to ensure fairness in resource allocation.

[0047] By formulating the above dynamic adjustment strategy, when the system performs execution adjustment, it uses intelligent optimization algorithms (such as genetic algorithms or simulated annealing algorithms) to globally optimize the allocation of traffic pool resources. The algorithm dynamically adjusts the resource allocation strategy according to the actual needs and prediction results of users to improve resource utilization.

[0048] Furthermore, the system adds a real-time alarm and intervention module. This module monitors traffic usage in real time by setting abnormal detection rules. Once an abnormal situation (such as sudden increase in traffic, abnormal behavior pattern, etc.) is found, the alarm mechanism is immediately triggered. The alarm information is notified to the administrator or user through various means such as email, SMS, system message, etc., and rapid response and intervention measures (such as limiting traffic, suspending service, etc.) are provided to ensure the legality and rationality of traffic usage.

[0049] In order to improve the practicality of the system, the system provides multi-dimensional and multi-level monitoring views, presenting key indicators such as traffic usage, user behavior, and system performance to administrators and users in an intuitive and easy-to-understand manner. Through visual display, monitoring efficiency and user experience are improved.

[0050] The intelligent dynamic adjustment method of traffic data and its corresponding system introduced in the above embodiments, in terms of social effects, the application of the disclosed intelligent dynamic adjustment algorithm helps to build a more fair and efficient traffic market environment and promote the healthy development of the information and communication industry. By reducing resource waste and environmental pollution, it promotes sustainable social development; in terms of economic effects, for operators, the algorithm can significantly improve the utilization rate and profitability of traffic pool resources and reduce operating costs. At the same time, by providing personalized services, it increases user stickiness and expands revenue sources. According to statistics, companies that adopt this algorithm can save an average of about 15% of operating costs and increase revenue by about 20% each year; in terms of technical effects, it not only promotes the advancement of traffic management technology, but also promotes the development of related fields such as machine learning and big data processing. The successful application of the disclosed intelligent dynamic adjustment algorithm provides a useful reference and reference for intelligent management in other fields.

[0051] In summary, the intelligent dynamic adjustment method of traffic data disclosed in this embodiment uses advanced machine learning algorithms (such as LSTM, ARIMA, etc.) to build a high-precision and high-stability traffic prediction model. The model can comprehensively consider multi-dimensional information such as historical data, time factors, user behavior, etc. to achieve accurate prediction of future traffic demand. In addition, the model must also have self-learning capabilities and be able to continuously optimize the prediction accuracy based on new data. A variety of machine learning algorithms (such as LSTM for processing time series data and ARIMA for seasonal trend prediction) are organically combined, and through weight allocation and model optimization, traffic prediction with higher accuracy than a single model is achieved. This fusion technology not only improves the accuracy of the prediction, but also enhances the adaptability and stability of the model.

[0052] And according to the prediction results and the current traffic pool resource status, an efficient dynamic adjustment strategy is designed. The strategy needs to comprehensively consider multiple factors such as user priority, demand urgency, historical consumption records, etc., and calculate the optimal traffic allocation plan through optimization algorithms (such as genetic algorithms and simulated annealing algorithms). At the same time, the execution of the strategy must ensure real-time, accuracy and stability to meet the needs of traffic pool management. The dynamic adjustment strategy not only relies on simple mathematical optimization algorithms, but also deeply integrates big data analysis technology. Through the comprehensive analysis of massive user data, historical traffic data and real-time market data, the dynamic adjustment strategy can more accurately identify user needs and market changes, thereby formulating a more scientific and reasonable traffic allocation plan.

[0053] Secondly, a multi-dimensional data analysis function is introduced to conduct a comprehensive analysis of the user's traffic usage behavior, geographic location, time preference and other information to provide users with personalized traffic management services. This service can not only improve user satisfaction and loyalty, but also bring more business opportunities and revenue sources to enterprises. The service can automatically match the most suitable traffic management solution based on the user's historical behavior, preferences and current needs. The intelligent matching algorithm continuously optimizes the matching logic by continuously learning from the user's feedback and behavior data, so that the services provided are closer to the actual needs of users. The intelligent matching technology of this personalized service algorithm is a key technical innovation of the present invention in improving user experience and increasing user stickiness.

[0054] And build an efficient real-time monitoring and rapid response mechanism to ensure that abnormal situations in traffic usage can be discovered and handled in a timely manner. The mechanism must have high sensitivity, high accuracy and high reliability, and be able to quickly trigger alarms and notify relevant personnel to handle them, so as to reduce potential economic losses and social impacts. This mechanism not only has efficient data processing capabilities, but also has built-in innovative anomaly detection and response mechanisms. By combining machine learning algorithms and rule engines, this mechanism can analyze traffic usage data in real time, quickly identify abnormal behaviors and trigger corresponding alarms and processing processes. In addition, the system also has the ability to automatically adjust monitoring thresholds and optimize alarm strategies to cope with the ever-changing traffic environment and user needs.

[0055] At the same time, in the corresponding system architecture, the principles of modularization and scalability can be adopted to make each functional module both independent and closely coordinated. This design not only improves the flexibility and maintainability of the system, but also enables the system to easily cope with new business needs and technical challenges that may arise in the future. By adding new modules or upgrading existing modules, the system can continuously expand its functions and application scope and maintain its technological leadership.

[0056] As a feasible implementation method of this embodiment, a rule-based approach can be adopted to adjust the traffic distribution, relying on a preset rule set to adjust the traffic resource allocation. By setting a series of rules based on historical data or expert experience, the system will match the preset rules according to the current traffic usage and execute corresponding adjustment measures. For example, when traffic usage exceeds a certain threshold, the resource quota is automatically increased or the traffic allocation during non-peak hours is reduced. This rule-based approach is simple to implement and does not require complex algorithm support, but the adjustment strategy is relatively fixed, lacks flexibility and predictability, and is difficult to cope with a rapidly changing market environment. Compared with the aforementioned preferred embodiment of the intelligent dynamic adjustment method for traffic data disclosed above, the prediction accuracy and resource utilization are lower.

[0057] Secondly, in the acquisition of the machine learning model, a simplified version of the model can be used, that is, a relatively simple machine learning model (such as linear regression, decision tree, etc.) can be used for traffic prediction. Although these models are not as complex as deep learning models (such as LSTM), they can also capture the trends and patterns of traffic usage to a certain extent, thereby supporting a certain degree of dynamic adjustment. Although this simplified version of the model can improve the accuracy of traffic prediction and the efficiency of resource adjustment to a certain extent, compared with the complex machine learning model disclosed in the aforementioned preferred embodiment of the intelligent dynamic adjustment method for traffic data, its prediction accuracy and dynamic adjustment capabilities may be lacking, especially when processing complex and changeable traffic data.

[0058] In terms of resource allocation, a feasible way is to adopt a dynamic resource allocation service based on cloud services, which uses the dynamic resource allocation service provided by the existing cloud service provider to transmit traffic data to the cloud service end in real time through the API interface. The cloud server performs data processing and predictive analysis, and returns adjustment suggestions. The system performs dynamic adjustment of traffic resources according to the suggestions returned by the cloud. This dynamic resource allocation service based on cloud services can achieve a certain degree of dynamic adjustment with the help of the powerful data processing capabilities and flexible resource allocation mechanism of cloud services. Compared with the resource allocation service disclosed in the embodiment of the aforementioned preferred intelligent dynamic adjustment method for traffic data, the degree of customization of its adjustment strategy and prediction model may be limited by the capabilities and service scope of the cloud service provider, making it difficult to achieve completely personalized optimization management.

[0059] Finally, in terms of the adjustment mechanism, a feasible approach is a semi-automatic adjustment mechanism driven by user feedback, which combines user feedback with an automatic adjustment mechanism. The system regularly collects user feedback on traffic usage, and combines basic data analysis to manually or semi-automatically adjust the traffic resource allocation strategy. For example, based on user complaints and satisfaction survey results, increase the traffic quota in a specific area or reduce the quota of users who abuse resources. This user feedback-driven semi-automatic adjustment mechanism enhances the pertinence and flexibility of the system adjustment strategy by introducing a user feedback mechanism. However, its adjustment speed and accuracy are limited by the timeliness and accuracy of user feedback, as well as the efficiency and accuracy of manual intervention. Compared with the technical means of the aforementioned preferred embodiment of the intelligent dynamic adjustment method for traffic data disclosed above, its degree of automation and real-time performance are lower.

[0060] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein or their combination. For software implementation, part or all of the flow of the embodiment can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein the communication medium includes any medium that is convenient for transmitting a computer program from one place to another. The storage medium can be any available medium that a computer can access. The computer-readable storage medium can include but is not limited to RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer.

[0061] Finally, it should be noted that the above is only a preferred embodiment of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent dynamic adjustment method for flow data, characterized in that: The method comprises the following steps in sequence: Data collection: Collect traffic usage data from various data sources; Data processing: pre-process the collected traffic usage data and extract key indicators for subsequent analysis; Traffic forecasting: Use the trained machine learning model to perform forecast analysis on the preprocessed data to obtain the traffic demand forecast for a period of time in the future; Dynamic adjustment strategy formulation: Based on the traffic prediction results and the current traffic pool resource status, the optimization algorithm is used to calculate the optimal traffic allocation plan; Execute adjustment: Automatically adjust the traffic pool resource allocation according to the established adjustment strategy, including increasing or decreasing the user's traffic quota; Real-time monitoring and alarm: Continuously monitor traffic usage, immediately trigger the alarm mechanism when abnormal situations are found, and notify relevant personnel to handle them.

2. The intelligent dynamic adjustment method of flow data according to claim 1 is characterized in that: In the data collection, the traffic usage data is collected from various data sources periodically or according to trigger conditions through distributed data collection nodes.

3. The intelligent dynamic adjustment method of flow data according to claim 1 is characterized in that: In the data processing, the preprocessing includes cleaning, deduplication and integration.

4. The intelligent dynamic adjustment method of flow data according to claim 1 is characterized in that: In the traffic prediction, the training steps of the machine learning model are: using a machine learning algorithm to build a traffic prediction model, training historical traffic data, and generating a machine learning model that predicts future traffic demand.

5. The intelligent dynamic adjustment method of flow data according to claim 2 is characterized in that: When training the machine learning model, time, user behavior patterns, network conditions and corresponding historical traffic data are also included as model training data.

6. The intelligent dynamic adjustment method of flow data according to claim 1, characterized in that: The optimization algorithm is a genetic algorithm or a simulated annealing algorithm.

7. The intelligent dynamic adjustment method of flow data according to claim 1, characterized in that: In the real-time monitoring and alarm, abnormality detection is performed on the traffic usage according to the set abnormality detection rules, and the abnormal situation includes limiting the traffic or suspending the service.

8. The intelligent dynamic adjustment method of flow data according to claim 1, characterized in that: The intelligent dynamic adjustment method of traffic data also includes: Traffic usage, user behavior, and operating performance are displayed through monitoring views.

Images