Network dot network health monitoring method, device and equipment and storage medium

Abstract

The application relates to the technical field of computers and discloses a network point network health monitoring method, which comprises the following steps: permission identification and evaluation, receiving a user login request, identifying the role permission level of the user, and evaluating the data access mode of the user; data screening and desensitization, screening the corresponding network data range according to the role permission level, and performing network data desensitization and access control strategy; abnormality detection and grading, performing state filtering on the screened network point data, identifying abnormal data, and generating an early warning grade; visualization and business analysis, marking the life cycle stage of each network point in a visual chart, and calculating a network point shipment rate index in multiple time dimensions; health degree evaluation and display, displaying the score, early warning and prediction results; feedback collection and optimization, collecting feedback data of the user on the early warning and score results. The application can safely control, accurately warn and integrate business, quantitatively evaluate and close-loop optimize, and improve the network point monitoring safety and operation and maintenance efficiency.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, device, and storage medium for monitoring the health of a network. Background Technology

[0002] In distributed network architectures and multi-site business operation scenarios, network health monitoring is a core component for ensuring business continuity. Existing monitoring systems still have significant limitations in terms of granularity and access control when dealing with multi-level operation and maintenance roles and massive amounts of heterogeneous site data.

[0003] First, existing systems mostly employ fixed permission models based on static roles, lacking dynamic analysis of actual user behavior characteristics. This makes them prone to unauthorized access. Protection methods for confidential data are simplistic, failing to balance data security and availability. Second, anomaly detection relies heavily on preset, single, static thresholds, resulting in coarse-grained alert levels. This model struggles to adapt to dynamic changes in network size, business hours, and lifecycles, leading to frequent false alarms and missed alarms. This makes it difficult for maintenance personnel to prioritize tasks, impacting emergency response efficiency. Furthermore, existing solutions primarily focus on purely technical indicators of network devices and links, failing to deeply correlate and integrate them with network site business attributes. This results in a disconnect between network monitoring data and network site operations, hindering business decision-making.

[0004] Therefore, it is necessary to invent a method, device, equipment, and storage medium for monitoring the health of branch networks that can control access levels and access behaviors, improve the security of confidential data in the system, accurately identify abnormal situations, and provide support for branch business decisions. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and storage medium for monitoring the health of a branch network, which is used to control access levels and access behaviors, improve the security of confidential data in the system, accurately identify abnormal situations, and provide support for branch business decisions.

[0006] The first aspect of this invention provides a method for monitoring the health of a network at a business location, comprising: Permission identification and assessment: Receive user login requests, identify the user's role and permission level, and assess the user's data access patterns; Data filtering and desensitization: Based on the role permission level, filter the corresponding network data range, and desensitize and implement access control policies for the network data; Anomaly detection and rating: The system filters the status of the selected network data, identifies abnormal data, and generates early warning ratings. Visualization and business analysis generate visual charts of the network structure, mark the life cycle stage of each network in the visual charts, and calculate the network shipment rate indicators in multiple time dimensions. Health assessment and display: Calculate the comprehensive health score of the network outlets and display the score, early warning and prediction results; Feedback collection and optimization: Collect user feedback data on warnings and scoring results, and optimize the model parameters.

[0007] Optionally, in a first implementation of the first aspect of the present invention, the permission identification and evaluation, which involves receiving a user login request, identifying the user's role permission level, and evaluating the user's data access pattern, includes: Receive user login information, which includes at least username, password, login terminal IP address and login timestamp, and perform format validation on the login information to filter invalid requests; User roles are matched based on the RBAC model, and the user roles include at least super administrator, network administrator, ordinary operation and maintenance personnel, and read-only viewer; Collect users' historical login data and historical access records, compare them with the role's standard access patterns through a behavioral feature analysis model, evaluate the rationality of the current access pattern, and mark abnormal access behaviors.

[0008] Optionally, in a second implementation of the first aspect of the present invention, the data filtering and desensitization, based on the role permission level, filters the corresponding network data range and implements network data desensitization and access control strategies, including: Filter network data according to user role and permission levels; The core IP addresses of network outlets, device serial numbers, contact information of maintenance personnel, and sensitive business data contained in the filtered data are anonymized. Implement restrictive measures for abnormal access behavior. These measures include at least restricting the scope of access, reducing operating privileges, requiring secondary verification, and recording all data access operation logs.

[0009] Optionally, in a third implementation of the first aspect of the present invention, the anomaly detection and rating involves performing status filtering on the screened network point data, identifying abnormal data, and generating an early warning rating, including: Valid outlets are selected according to preset criteria; Abnormal data is identified through data quality detection algorithms, and the abnormal data is then labeled, classified, and processed. A multi-level anomaly detection algorithm is used for detection, and an early warning rating is generated based on the detection results. The early warning rating corresponds to different handling priorities.

[0010] Optionally, in the fourth implementation of the first aspect of the present invention, the visualization and business analysis generates a visual chart of the network structure, marks the lifecycle stage of each network point in the visual chart, and calculates network shipment rate indicators across multiple time dimensions, including: Extract core information from branch offices to generate visual charts; The life cycle stage is evaluated based on the historical operation data of the network, and the life cycle stage includes at least the initialization stage, growth stage, stable operation stage, decline stage and elimination stage. The system calculates the shipment rate of individual outlets and the overall shipment rate at daily, weekly, monthly, and quarterly levels. Combined with auxiliary indicators such as year-on-year / month-on-month growth rates, it uses an LSTM model to generate future cycle shipment trend predictions.

[0011] Optionally, in the fifth implementation of the first aspect of the present invention, the health assessment and display, which calculates a comprehensive health score for the network outlets and displays the score, warning, and prediction results, includes: Construct a health assessment index system for outlets and assign corresponding weights to each index; The outlets are classified into different levels based on their overall health score; The system displays health scores, grading results, weak indicators, optimization suggestions, early warning information, and shipment trend forecasts.

[0012] Optionally, in a sixth implementation of the first aspect of the present invention, the feedback collection and optimization involves collecting user feedback data on warning and scoring results, and optimizing the model parameters, including: Collect user feedback, and the feedback types include at least false alarms, missed alarms, unreasonable scores, and prediction bias. Effective feedback data is input into the online learning mechanism to optimize the parameters of the behavioral feature analysis model, data quality detection algorithm, multi-level anomaly detection algorithm, time series prediction model, and multi-index fusion evaluation model. The optimization effect was verified by comparing the early warning accuracy, scoring rationality, and prediction deviation rate before and after optimization, and the optimized model parameters were applied to the monitoring process.

[0013] A second aspect of the present invention provides a network health monitoring device for service outlets, comprising: The permission identification and evaluation module is used to receive user login requests, identify the user's role permission level, and evaluate the user's data access pattern. The permission identification and evaluation module includes: The receiving unit is used to receive user login information, which includes at least a username, password, login terminal IP address and login timestamp, and performs format verification on the login information to filter invalid requests. The matching unit is used to match user roles based on the RBAC model. The user roles include at least super administrator, network administrator, ordinary operation and maintenance personnel, and read-only viewer. The first evaluation unit is used to collect users' historical login data and historical data access records. By comparing the behavioral feature analysis model with the standard access pattern of the role, it evaluates the rationality of the current access pattern and marks abnormal access behavior.

[0014] The data filtering and desensitization module is used to filter the corresponding range of network data according to the role permission level, and to desensitize and implement access control policies for the network data. The data filtering and desensitization module includes: The first filtering unit is used to filter network data according to user role and permission levels; The processing unit is used to de-identify the core IP addresses of network points, device serial numbers, contact information of maintenance personnel, and sensitive business data contained in the filtered data. The execution unit is used to implement restrictive measures for abnormal access behavior. The restrictive measures include at least restricting the access scope, reducing operation privileges, requiring secondary verification, and recording all data access operation logs.

[0015] The anomaly detection and rating module is used to filter the status of the screened network point data, identify abnormal data, and generate early warning ratings. The anomaly detection and rating module includes: The second screening unit is used to screen valid outlets according to preset criteria; The labeling unit is used to identify anomalous data through data quality detection algorithms, and to label, classify, and process the anomalous data. The early warning unit is used to perform detection using a multi-level anomaly detection algorithm and generate an early warning rating based on the detection results. The early warning rating corresponds to different handling priorities.

[0016] The visualization and business analysis module is used to generate visual charts of the network structure, mark the life cycle stage of each network in the visual charts, and calculate the network shipment rate indicators in multiple time dimensions. The visualization and business analysis module includes: The extraction unit is used to extract core information about the network points and generate visual charts. The second evaluation unit is used to evaluate the life cycle stage based on the historical operation data of the network. The life cycle stage includes at least the initialization stage, growth stage, stable operation stage, decline stage and elimination stage. The calculation unit is used to calculate the shipment rate of a single outlet and the overall shipment rate in the daily, weekly, monthly and quarterly dimensions. Combined with auxiliary indicators such as year-on-year / month-on-month growth rates, it uses an LSTM model to generate future cycle shipment trend predictions.

[0017] The health assessment and display module is used to calculate the comprehensive health score of the network outlets and display the score, warning and prediction results; The health assessment and display module includes: The building unit is used to construct a health assessment indicator system for outlets and assign corresponding weights to each indicator. The division unit is used to classify outlets into different levels based on a comprehensive health score. The display unit is used to show the health score, grading results, weak indicators, optimization suggestions, early warning information and shipment trend forecast.

[0018] The feedback collection and optimization module is used to collect user feedback data on warning and scoring results and optimize the model parameters.

[0019] The feedback collection and optimization module includes: A collection unit is used to collect user feedback, the feedback types of which include at least false alarms, missed alarms, unreasonable ratings, and prediction biases. The optimization unit is used to input effective feedback data into the online learning mechanism to optimize the parameters of the behavioral feature analysis model, data quality detection algorithm, multi-level anomaly detection algorithm, time series prediction model, and multi-index fusion evaluation model. The application unit is used to verify the optimization effect by comparing the early warning accuracy, scoring rationality, and prediction deviation rate before and after optimization, and to apply the optimized model parameters to the monitoring process.

[0020] A third aspect of the present invention provides a network health monitoring device for a business site, comprising a memory and at least one processor, wherein the memory stores computer-readable instructions; The at least one processor invokes the computer-readable instructions in the memory to perform the various steps of the site network health monitoring method as described above.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the various steps of the network health monitoring method described above.

[0022] This invention ensures access security through dynamic permission control and data anonymization, employs multi-level algorithms to accurately detect anomalies and provide tiered early warnings, and deeply integrates network monitoring with business indicators such as branch lifecycle and shipment rate to achieve quantifiable health assessment and visualization. Furthermore, it optimizes model parameters based on user feedback in a closed loop, effectively improving system security, anomaly identification accuracy, and operational decision-making efficiency. It is suitable for long-term stable use in distributed multi-branch scenarios. Attached Figure Description

[0023] Figure 1 A flowchart illustrating the network health monitoring method for service outlets provided in an embodiment of the present invention; Figure 2 A flowchart of the permission identification and evaluation steps provided in this embodiment of the invention; Figure 3 A flowchart of the data filtering and desensitization steps provided in this embodiment of the invention; Figure 4 A flowchart of the anomaly detection and rating steps provided in this embodiment of the invention; Figure 5 A flowchart of the visualization and business analysis steps provided in the embodiments of the present invention; Figure 6 A flowchart illustrating the health assessment and display steps provided in this embodiment of the invention; Figure 7 This is a schematic diagram of the network health monitoring device for service outlets provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the network health monitoring device for service outlets provided in an embodiment of the present invention. Detailed Implementation

[0024] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 The first embodiment of the network health monitoring method for service outlets in this invention includes: S101. Permission identification and evaluation: Receive user login request, identify the user's role and permission level, and evaluate the user's data access pattern. S102. Data filtering and desensitization: Based on the role permission level, filter the corresponding network data range, and desensitize and implement access control policies for the network data. S103, Anomaly Detection and Rating: The selected network point data is filtered by status to identify abnormal data and generate early warning ratings. S104. Visualization and Business Analysis: Generate a visual chart of the network structure, mark the life cycle stage of each network in the visual chart, and calculate the network shipment rate index in multiple time dimensions. S105. Health Assessment and Display: Calculate the comprehensive health score of the network outlets and display the score, early warning and prediction results. S106. Feedback Collection and Optimization: Collect user feedback data on warning and scoring results, and optimize the model parameters.

[0026] This embodiment achieves refined management of user permissions and access behavior through a closed-loop design encompassing access control, data security, anomaly detection, business analysis, health assessment, and feedback optimization. This ensures the security of sensitive data, accurately identifies branch network anomalies and provides tiered early warnings, and combines business indicators to complete visual analysis and quantitative health assessments. Furthermore, it can continuously optimize the model based on user feedback, effectively improving the security, accuracy, and intelligence level of branch network monitoring, and providing reliable support for operation and maintenance management and business decision-making.

[0027] Please see Figure 2 In a second embodiment of the network health monitoring method for service outlets in this invention, the permission identification and evaluation, which involves receiving a user login request, identifying the user's role and permission level, and evaluating the user's data access pattern, includes: S201. Receive user login information, the login information including at least username, password, login terminal IP address and login timestamp, and perform format verification on the login information to filter invalid requests; S202. Match user roles based on the RBAC model. The user roles include at least super administrator, network administrator, ordinary operation and maintenance personnel and read-only viewer. Clarify the network data access scope, operation permissions and data operation thresholds corresponding to each role. S203. Collect user's historical login data and historical data access records, compare them with the standard access patterns of the role through a behavioral feature analysis model, evaluate the rationality of the current access pattern, and mark abnormal access behavior.

[0028] This embodiment combines multi-dimensional login information verification, refined RBAC role matching, and dynamic evaluation of user access behavior to upgrade permission control from static allocation to dynamic verification. It can effectively filter illegal requests, clarify the operational boundaries of different roles, and promptly detect abnormal access through historical behavior comparison. This greatly improves the permission security and access control accuracy of the branch network system, laying a solid security foundation for subsequent data access and business operations.

[0029] Please see Figure 3 In a third embodiment of the network health monitoring method for service points in this invention, the data filtering and desensitization, based on the role and permission levels, filters the corresponding network data range and implements desensitization and access control policies for the network data, including: S301. Filter network data according to user role and permission levels. Super administrators can access all network data and core operation and maintenance data, ordinary operation and maintenance personnel can only access basic data of network points within their jurisdiction, and read-only users can only access network status summary data. S302. De-identify the core IP addresses of network points, device serial numbers, contact information of maintenance personnel, and sensitive business data contained in the filtered data; S303. Implement restrictive measures for abnormal access behavior. The restrictive measures include at least limiting the access scope, reducing operation privileges, requiring secondary verification, and recording all data access operation logs.

[0030] This embodiment implements differentiated data filtering based on permission levels, de-identifies and protects core sensitive information, dynamically restricts abnormal access, and retains operation logs throughout the process. This not only strictly ensures that the core data and business information of the branch are not leaked or used without authorization, but also provides suitable data services for different roles under the premise of security and compliance, effectively improving the system's data security management capabilities and access compliance.

[0031] Please see Figure 4 In the fourth embodiment of the network health monitoring method for service outlets in this invention, the anomaly detection and rating involves filtering the selected service outlet data by status, identifying abnormal data, and generating an early warning rating, including: S401. Select valid network points according to preset standards. The criteria for determining valid network points include at least the following: normal online status of equipment, data transmission 24 hours a day, stable equipment connection, and no hardware fault alarms. Exclude invalid network points. S402. Identify abnormal data from the dimensions of data integrity, accuracy and consistency through data quality detection algorithms. After marking and classifying abnormal data, process it by completing, correcting or removing it. S403. A multi-level anomaly detection algorithm is adopted, wherein the multi-level includes a basic detection layer, a deep detection layer and a correlation detection layer. Based on the detection results, a five-level early warning rating is generated, and the five-level early warning rating corresponds to different handling priorities and preliminary handling suggestions.

[0032] This embodiment combines precise screening of effective service points, multi-dimensional data quality inspection, and multi-level anomaly detection. It first eliminates invalid service points and abnormal data, and then comprehensively identifies faults from three levels: basic, in-depth, and correlation. At the same time, it generates five-level early warnings and corresponding handling suggestions, which effectively reduces false alarms and missed alarms, improves the accuracy of anomaly identification, and makes operation and maintenance handling more efficient and targeted.

[0033] Please see Figure 5 The fifth embodiment of the network health monitoring method for service outlets in this invention includes visualization and business analysis, which generates a visual chart of the service outlet structure, marks the lifecycle stage of each service outlet in the visual chart, and calculates multi-time-dimensional service outlet shipment rate indicators, including: S501. Extract core information of network points to generate visualization charts. The visualization chart types include at least topology diagrams, heat maps, and list diagrams. The core information includes at least network point ID, geographical location, equipment configuration, network topology relationship, and current operating status. S502. Evaluate the life cycle stage based on the historical operation data of the network. The historical operation data includes at least the running time, equipment wear and tear, data transmission stability, maintenance frequency and failure rate. The life cycle stage includes at least the initialization stage, growth stage, stable operation stage, decline stage and retirement stage. S503 calculates the shipment rate of a single outlet and the overall shipment rate in the daily, weekly, monthly, and quarterly dimensions. Combined with auxiliary indicators such as year-on-year / month-on-month growth rates, it uses the ARIMA model or LSTM model to generate a forecast of future cycle shipment trends.

[0034] This embodiment uses various types of visual charts to intuitively present the network status of outlets, accurately divides the life cycle of outlets by combining operational data, and conducts business analysis based on multi-time dimension shipment rate and time series prediction models. This achieves deep integration of technical monitoring and business operations, which not only makes it easier for maintenance personnel to quickly grasp the overall situation, but also provides reliable data support for outlet layout optimization, resource allocation and business planning.

[0035] Please see Figure 6 The sixth embodiment of the network health monitoring method for service points in this invention includes a health assessment and display process that calculates a comprehensive health score for the service points and displays the score, warning, and prediction results, including: S601. Construct a health assessment index system for network outlets. The index system shall include at least the network outlet operation status, data quality, shipping capacity, life cycle stage and early warning rating, and assign corresponding weights to each index. S602. Based on the comprehensive health score, the outlets are divided into four levels: excellent, good, qualified and unqualified. Excellent corresponds to 90-100 points, good corresponds to 80-89 points, qualified corresponds to 60-79 points, and unqualified corresponds to below 60 points. S603 pushes health scores, grading results, weak indicators, optimization suggestions, early warning information, and shipment trend predictions to the network management dashboard, supporting interactive operations such as viewing network details, filtering specific levels of networks, and comparing historical data.

[0036] Furthermore, the feedback collection and optimization involves collecting user feedback data on warnings and scoring results, and optimizing the model parameters, including: User feedback is collected through feedback buttons on the network management dashboard or feedback forms in the background. The feedback types include at least false alarms, missed alarms, unreasonable scores, and prediction deviations. Effective feedback data is input into the online learning mechanism to optimize the parameters of the behavioral feature analysis model, data quality detection algorithm, multi-level anomaly detection algorithm, time series prediction model, and multi-index fusion evaluation model. The optimization effect was verified by comparing the early warning accuracy, scoring rationality, and prediction deviation rate before and after optimization, and the optimized model parameters were applied to the monitoring process.

[0037] This embodiment constructs a multi-dimensional health assessment system and performs quantitative scoring and level classification, transforming the network operation status into intuitive and quantifiable results. Combined with the network management dashboard, it achieves visualization and interactive display. At the same time, based on user feedback, it establishes an online learning and model optimization mechanism, which can continuously improve the accuracy of detection, assessment and prediction, making operation and maintenance management more intuitive and decision-making more scientific, and realizing the self-improvement and long-term stable operation of the monitoring system.

[0038] The network health monitoring method for service points in the embodiments of the present invention has been described above. The apparatus in the embodiments of the present invention will be described below. Please refer to [link / reference]. Figure 7 The implementation method of the network health monitoring device for service outlets in this invention includes: The permission identification and evaluation module 701 is used to receive user login requests, identify the user's role permission level, and evaluate the user's data access pattern. In some embodiments, the permission identification and evaluation module 701 includes: The receiving unit 7011 is used to receive user login information, which includes at least a username, password, login terminal IP address and login timestamp, and performs format validation on the login information to filter invalid requests. Matching unit 7012 is used to match user roles based on the RBAC model, wherein the user roles include at least super administrator, network administrator, ordinary operation and maintenance personnel and read-only viewer; The first evaluation unit 7013 is used to collect users' historical login data and historical data access records, compare them with the standard access patterns of the role through a behavioral feature analysis model, evaluate the rationality of the current access pattern, and mark abnormal access behavior.

[0039] This embodiment combines multi-dimensional login information verification, refined RBAC role matching, and dynamic evaluation of user access behavior to upgrade permission control from static allocation to dynamic verification. It can effectively filter illegal requests, clarify the operational boundaries of different roles, and promptly detect abnormal access through historical behavior comparison. This greatly improves the permission security and access control accuracy of the branch network system, laying a solid security foundation for subsequent data access and business operations.

[0040] The data filtering and desensitization module 702 is used to filter the corresponding network data range according to the role permission level, and to desensitize and implement access control policies for the network data. In some embodiments, the data filtering and desensitization module 702 includes: The first filtering unit 7021 is used to filter network data according to user role and permission levels; Processing unit 7022 is used to de-identify the core IP address of the network point, the serial number of the device, the contact information of the operation and maintenance personnel and sensitive business data contained in the filtered data; The execution unit 7023 is used to implement restrictive measures for abnormal access behavior. The restrictive measures include at least restricting the access scope, reducing operation privileges, and requiring secondary verification, and recording all data access operation logs.

[0041] This embodiment implements differentiated data filtering based on permission levels, de-identifies and protects core sensitive information, dynamically restricts abnormal access, and retains operation logs throughout the process. This not only strictly ensures that the core data and business information of the branch are not leaked or used without authorization, but also provides suitable data services for different roles under the premise of security and compliance, effectively improving the system's data security management capabilities and access compliance.

[0042] The anomaly detection and rating module 703 is used to perform status filtering on the screened network point data, identify abnormal data, and generate early warning ratings. In some embodiments, the anomaly detection and rating module 703 includes: The second screening unit 7031 is used to screen valid outlets according to preset criteria; The labeling unit 7032 is used to identify abnormal data through a data quality detection algorithm, label and classify the abnormal data, and process it. The early warning unit 7033 is used to perform detection using a multi-level anomaly detection algorithm and generate an early warning rating based on the detection results. The early warning rating corresponds to different handling priorities.

[0043] This embodiment combines precise screening of effective service points, multi-dimensional data quality inspection, and multi-level anomaly detection. It first eliminates invalid service points and abnormal data, and then comprehensively identifies faults from three levels: basic, in-depth, and correlation. At the same time, it generates five-level early warnings and corresponding handling suggestions, which effectively reduces false alarms and missed alarms, improves the accuracy of anomaly identification, and makes operation and maintenance handling more efficient and targeted.

[0044] The Visualization and Business Analysis Module 704 is used to generate visual charts of the network structure, mark the life cycle stage of each network in the visual charts, and calculate the network shipment rate indicators in multiple time dimensions. In some embodiments, the visualization and business analysis module 704 includes: Extraction unit 7041 is used to extract core information of network points and generate visual charts; The second evaluation unit 7042 is used to evaluate the life cycle stage based on the historical operation data of the network. The life cycle stage includes at least the initialization stage, growth stage, stable operation stage, decline stage and elimination stage. The calculation unit 7043 is used to calculate the single-site shipment rate and overall shipment rate in the daily, weekly, monthly and quarterly dimensions. Combined with auxiliary indicators such as year-on-year / month-on-month growth rate, it uses an LSTM model to generate future cycle shipment trend predictions.

[0045] This embodiment uses various types of visual charts to intuitively present the network status of outlets, accurately divides the life cycle of outlets by combining operational data, and conducts business analysis based on multi-time dimension shipment rate and time series prediction models. This achieves deep integration of technical monitoring and business operations, which not only makes it easier for maintenance personnel to quickly grasp the overall situation, but also provides reliable data support for outlet layout optimization, resource allocation and business planning.

[0046] The health assessment and display module 705 is used to calculate the comprehensive health score of the network outlets and display the score, early warning and prediction results; In some embodiments, the health assessment and display module 705 includes: Building unit 7051 is used to build a health assessment index system for outlets and assign corresponding weights to each index. Division unit 7052 is used to classify outlets into different levels based on a comprehensive health score; Display unit 7053 is used to display health scores, grading results, weak indicators, optimization suggestions, early warning information and shipment trend forecasts.

[0047] The feedback collection and optimization module 706 is used to collect user feedback data on warning and scoring results and optimize the model parameters.

[0048] In some embodiments, the feedback collection and optimization module 706 includes: The collection unit 7061 is used to collect user feedback, and the feedback types include at least false alarms, missed alarms, unreasonable scores, and prediction deviations. The optimization unit 7062 is used to input effective feedback data into the online learning mechanism to optimize the parameters of the behavioral feature analysis model, data quality detection algorithm, multi-level anomaly detection algorithm, time series prediction model and multi-index fusion evaluation model. Application unit 7063 is used to verify the optimization effect by comparing the early warning accuracy, scoring rationality and prediction deviation rate before and after optimization, and to apply the optimized model parameters to the monitoring process.

[0049] This embodiment constructs a multi-dimensional health assessment system and performs quantitative scoring and level classification, transforming the network operation status into intuitive and quantifiable results. Combined with the network management dashboard, it achieves visualization and interactive display. At the same time, based on user feedback, it establishes an online learning and model optimization mechanism, which can continuously improve the accuracy of detection, assessment and prediction, making operation and maintenance management more intuitive and decision-making more scientific, and realizing the self-improvement and long-term stable operation of the monitoring system.

[0050] Figure 7 The structure of the network health monitoring device shown does not constitute a limitation on the network health monitoring device, and can implement the steps of the network health monitoring methods provided in the above-described method embodiments.

[0051] above Figure 7 The network health monitoring device of the present invention will be described in detail from the perspective of modular functional entities. The network health monitoring equipment of the present invention will be described in detail from the perspective of hardware processing.

[0052] Figure 8This is a schematic diagram of the structure of a network health monitoring device 800 provided in an embodiment of the present invention. The device 800 can vary significantly depending on its configuration and performance, and may include one or more central processing units (CPUs) 810 (e.g., one or more processors) and a memory 820, and one or more storage media 830 (e.g., one or more mass storage devices) for storing application programs 833 or data 832. The memory 820 and storage media 830 can be temporary or persistent storage. The program stored in the storage media 830 may include one or more modules (not shown), each module including a series of instruction operations on the device 800. Furthermore, the processor 810 may be configured to communicate with the storage media 830 and execute the series of instruction operations in the storage media on the device 800.

[0053] Device 800 may also include one or more power supplies 840, one or more wired or wireless network interfaces 850, one or more input / output interfaces 860, and / or one or more operating systems 831, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc.

[0054] This invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the network health monitoring method.

[0055] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0056] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0057] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring the health of a network of service outlets, characterized in that, include: Permission identification and assessment: Receive user login requests, identify the user's role and permission level, and assess the user's data access patterns; Data filtering and desensitization: Based on the role permission level, filter the corresponding network data range, and desensitize and implement access control policies for the network data; Anomaly detection and rating: The system filters the status of the selected network data, identifies abnormal data, and generates early warning ratings. Visualization and business analysis generate visual charts of the network structure, mark the life cycle stage of each network in the visual charts, and calculate the network shipment rate indicators in multiple time dimensions. Health assessment and display: Calculate the comprehensive health score of the network outlets and display the score, early warning and prediction results; Feedback collection and optimization: Collect user feedback data on warnings and scoring results, and optimize the model parameters.

2. The network health monitoring method for service outlets according to claim 1, characterized in that, The permission identification and evaluation process involves receiving user login requests, identifying the user's role and permission levels, and evaluating the user's data access patterns, including: Receive user login information, which includes at least username, password, login terminal IP address and login timestamp, and perform format validation on the login information to filter invalid requests; User roles are matched based on the RBAC model, and the user roles include at least super administrator, network administrator, ordinary operation and maintenance personnel, and read-only viewer; Collect users' historical login data and historical access records, compare them with the standard access patterns of the role through a behavioral feature analysis model, evaluate the rationality of the current access pattern, and mark abnormal access behavior.

3. The network health monitoring method for service outlets according to claim 2, characterized in that, The data filtering and desensitization, based on the role permission level, filters the corresponding range of network data and implements desensitization and access control policies for the network data, including: Filter network data according to user role and permission levels; The core IP addresses of network outlets, device serial numbers, contact information of maintenance personnel, and sensitive business data contained in the filtered data are anonymized. Implement restrictive measures for abnormal access behavior. These measures include at least restricting the scope of access, reducing operating privileges, requiring secondary verification, and recording all data access operation logs.

4. The network health monitoring method for service outlets according to claim 3, characterized in that, The anomaly detection and rating process filters the selected branch data by status, identifies abnormal data, and generates early warning ratings, including: Valid outlets are selected according to preset criteria; Abnormal data is identified through data quality detection algorithms, and the abnormal data is then labeled, classified, and processed. A multi-level anomaly detection algorithm is used for detection, and an early warning rating is generated based on the detection results. The early warning rating corresponds to different handling priorities.

5. The network health monitoring method for service outlets according to claim 4, characterized in that, The visualization and business analysis generate a visual chart of the network structure, marking the lifecycle stage of each network point in the chart, and calculating network shipment rate indicators across multiple time dimensions, including: Extract core information from branch offices to generate visual charts; The life cycle stage is evaluated based on the historical operation data of the network, and the life cycle stage includes at least the initialization stage, growth stage, stable operation stage, decline stage and elimination stage. The system calculates the shipment rate of individual outlets and the overall shipment rate at daily, weekly, monthly, and quarterly levels. Combined with auxiliary indicators such as year-on-year / month-on-month growth rates, it uses an LSTM model to generate future cycle shipment trend predictions.

6. The network health monitoring method for service outlets according to claim 5, characterized in that, The health assessment and display process calculates a comprehensive health score for each branch and displays the score, warnings, and prediction results, including: Construct a health assessment index system for outlets and assign corresponding weights to each index; The outlets are classified into different levels based on their overall health score; The system displays health scores, grading results, weak indicators, optimization suggestions, early warning information, and shipment trend forecasts.

7. The network health monitoring method for service outlets according to claim 5, characterized in that, The feedback collection and optimization process involves collecting user feedback data on warnings and scoring results, and optimizing the model parameters, including: Collect user feedback, and the feedback types include at least false alarms, missed alarms, unreasonable scores, and prediction bias. Effective feedback data is input into the online learning mechanism to optimize the parameters of the behavioral feature analysis model, data quality detection algorithm, multi-level anomaly detection algorithm, time series prediction model, and multi-index fusion evaluation model. The optimization effect was verified by comparing the early warning accuracy, scoring rationality, and prediction deviation rate before and after optimization, and the optimized model parameters were applied to the monitoring process.

8. A network health monitoring device for business outlets, characterized in that, include: The permission identification and evaluation module is used to receive user login requests, identify the user's role permission level, and evaluate the user's data access pattern. The data filtering and desensitization module is used to filter the corresponding range of network data according to the role permission level, and to desensitize and implement access control policies for the network data. The anomaly detection and rating module is used to filter the status of the screened network point data, identify abnormal data, and generate early warning ratings. The visualization and business analysis module is used to generate visual charts of the network structure, mark the life cycle stage of each network in the visual charts, and calculate the network shipment rate indicators in multiple time dimensions. The health assessment and display module is used to calculate the comprehensive health score of the network outlets and display the score, warning and prediction results; The feedback collection and optimization module is used to collect user feedback data on warning and scoring results and optimize the model parameters.

9. A network health monitoring device for business outlets, characterized in that, It includes a memory and at least one processor, wherein the memory stores computer-readable instructions; The at least one processor invokes the computer-readable instructions in the memory to perform the steps of the site network health monitoring method as described in any one of claims 1-7.

10. A computer-readable storage medium storing computer-readable instructions thereon, characterized in that, When the computer-readable instructions are executed by a processor, they implement the various steps of the site network health monitoring method as described in any one of claims 1-7.

Images