Household guest resource intelligent checking method and device, and electronic equipment

Abstract

The application provides a household resource intelligent checking method and device and electronic equipment, wherein the method comprises: obtaining resource data of a plurality of network resource entities related to household customer business from different data sources; analyzing the resource data to obtain a mapping relationship between the network resource entities and resource identifiers; constructing a resource topology relationship between the network resource entities based on the mapping relationship, and determining a resource portrait based on the resource topology relationship; intelligently diagnosing the resource portrait to obtain a diagnosis result, and determining a checking result of the resource data based on the diagnosis result. The application can associate scattered entity information based on the mapping relationship between the network resource entities and different resource identifiers by obtaining the resource data from different data sources, and can construct an accurate resource topology relationship and form a resource portrait based on the mapping relationship. The application can intelligently diagnose based on the resource portrait, realize the automation and intelligentization of household resource checking, and greatly improve the accuracy and real-time performance of resource checking.

Description

Technical Field

[0001] This invention relates to the field of resource management technology, and in particular to a method, apparatus and electronic device for intelligent inventory of household resources. Background Technology

[0002] With the widespread adoption of fiber optic broadband networks, the scale of residential customer services is expanding rapidly. The various resource data involved in the network, including optical cables, splitters, ports, and user access information, are characterized by massive volume, heterogeneity, and dynamic changes. This type of resource data is typically stored centrally in the operator's own resource capability center system. The accuracy of this resource data is a fundamental prerequisite for ensuring efficient service activation, rapid fault location, continuous network optimization, and precise marketing.

[0003] Currently, the management and inventory of residential customer resources faces numerous practical challenges. This work still largely relies on maintenance personnel for on-site manual verification, data entry, and comparison, resulting in a heavy workload, lengthy operation cycles, and difficulty in adapting to the rapidly changing business needs of large-scale networks. The manual operation process itself carries a high risk of error, easily leading to inconsistencies between the resource data recorded by the system and the actual physical network status. This results in issues such as invalid resources that remain unused for extended periods, abnormal resources that are actually online but not recorded in the system, and missing data. These problems severely impact the first-time success rate of service provisioning.

[0004] Furthermore, the traditional management model relying on periodic centralized checks is insufficient to reflect the dynamic changes in network resource status in real time. For example, port occupancy status cannot be updated in a timely and accurate manner, leading to distorted statistical analysis results of resource utilization. When the system detects data anomalies, it often lacks efficient and intelligent diagnostic tools, making it difficult to locate the root cause of the problem in a timely and accurate manner and automatically trigger the corresponding data repair process, thus failing to form a complete and efficient management loop. Summary of the Invention

[0005] This invention provides a method, device, and electronic device for intelligent inventory of household customer resources, which solves the systemic defects of existing technologies that rely mainly on manual operation and periodic centralized operation mode for inventory and management of household customer resources, resulting in low efficiency, high error rate, inability to reflect dynamic changes of resources in real time, and lack of intelligent diagnosis and closed-loop repair capabilities when data is abnormal.

[0006] This invention provides a method for intelligent inventory of customer resources, comprising the following steps.

[0007] Resource data of multiple network resource entities related to home customer business are obtained from different data sources; wherein the same network resource entity has different resource identifiers in different data sources; The resource data is parsed to obtain the mapping relationship between the network resource entity and the resource identifier; Based on the mapping relationship, a resource topology relationship between the network resource entities is constructed to determine the resource profile of the resource data based on the resource topology relationship. Obtain the diagnostic results of the intelligent diagnosis of the resource profile; Based on the diagnostic results, the results of the resource data clearance are determined.

[0008] According to the present invention, a method for intelligent inventory of home and guest resources, wherein constructing the resource topology relationship between the network resource entities based on the mapping relationship includes: Obtain a user account, use the user account as the starting network resource entity, and create a link structure including the starting network resource entity; The starting network resource entity is used as the current network resource entity. The next-hop network resource entity of the current network resource entity is queried in the mapping relationship, and the next-hop network resource entity is added to the link structure. The next-hop network resource entity is updated to the current network resource entity until the current network resource entity is a preset core network port entity, and the link structure is used as the resource topology relationship.

[0009] According to the present invention, a method for intelligent inventory of residential resources, wherein determining the resource profile of the resource data based on the resource topology relationship includes: A graph structure is generated based on the resource topology; wherein, the nodes in the graph structure correspond one-to-one with the network resource entities, and the edges in the graph structure represent the connection relationships of the network resource entities; Obtain dynamic operating parameters related to the network resource entity; the dynamic operating parameters are determined based on a performance attribute table or an alarm attribute table. Based on the resource identifier of the network resource entity, the dynamic operating parameters are associated with the nodes in the graph structure to obtain the resource profile.

[0010] According to the present invention, a method for intelligent inventory of customer resources includes obtaining diagnostic results of intelligent diagnosis of the resource profile, which includes: Based on the rule engine, a deterministic check is performed on the resource profile to obtain deterministic anomaly records; Extract the profile features of the resource profile, input the profile features into the diagnostic model, and obtain the probabilistic anomaly alarm output by the diagnostic model; the diagnostic model is trained based on the sample resource profile and the label probabilistic anomaly alarm corresponding to the sample resource profile. The diagnostic result is determined based on the deterministic anomaly record and / or the probabilistic anomaly alarm.

[0011] According to the present invention, a method for intelligent inventory of customer resources, the step of extracting the profile features of the resource profile includes: Obtain statistical values ​​of the historical port traffic of the network resource entity within a preset period; Determine the alarm frequency of the network resource entity within the preset period; Obtain the operational status data of other network resource entities that belong to the same parent network resource entity as the network resource entity; Based on the statistical values, the alarm frequency, and the operating status data, the profile features are determined.

[0012] According to the present invention, a method for intelligent inventory of customer resources, wherein the step of performing deterministic verification of the resource profile based on a rule engine to obtain deterministic anomaly records includes: The consistency between the running status data of the network resource entity in the resource profile and the status data of the associated port of the network resource entity is checked to obtain the first record; Logical verification is performed on the configuration data related to network services in the resource profile to obtain a second record; Based on the first record and / or the second record, the deterministic anomalous record is determined.

[0013] According to the present invention, a method for intelligent inventory of customer resources, wherein determining the inventory result of the resource data based on the diagnostic result includes: Based on the diagnostic results, the treatment type will be determined; When the handling type is automated repair, the data repair process is triggered to obtain the investigation results; When the handling type is manual handling, a maintenance work order is generated, and the investigation result is determined based on the maintenance result of the maintenance work order.

[0014] According to the present invention, a method for intelligent inventory of residential customer resources, wherein when the handling type is manual handling, a maintenance work order is generated, including: When the handling type is manual handling, the pending tasks related to the diagnostic results are obtained, as well as the candidate information of each maintenance personnel; the candidate information includes the skill tags, real-time load and geographical location of each maintenance personnel. Based on the complexity of the task to be processed, as well as the skill tags, real-time load, and geographical location of each maintenance personnel, the target maintenance personnel are determined. A maintenance work order, including the task to be processed, is dispatched to the target maintenance personnel.

[0015] The present invention also provides a smart inventory device for household customer resources, comprising the following units: The acquisition unit is used to acquire resource data of multiple network resource entities related to home customer business from different data sources; wherein the same network resource entity has different resource identifiers in different data sources; The parsing unit is used to parse the resource data to obtain the mapping relationship between the network resource entity and the resource identifier; The construction unit is used to construct the resource topology relationship between the network resource entities based on the mapping relationship, so as to determine the resource profile of the resource data based on the resource topology relationship; A diagnostic unit is used to obtain the diagnostic results of intelligent diagnosis of the resource profile; The determining unit is used to determine the results of the inventory of the resource data based on the diagnostic results.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the intelligent inventory method for household resources as described above.

[0017] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent inventory method for household resources as described above.

[0018] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the intelligent inventory method for household resources as described above.

[0019] The present invention provides a method, apparatus, and electronic device for intelligent inventory of household customer resources. These methods acquire resource data related to household customer business from different data sources, wherein the same network resource entity has different resource identifiers in different data sources. The resource data is parsed to obtain a mapping relationship between network resource entities and resource identifiers. Based on the mapping relationship, a resource topology relationship is constructed between network resource entities, and a resource profile of the resource data is determined based on the resource topology relationship. Intelligent diagnosis is performed on the resource profile to obtain diagnostic results, and the inventory results of the resource data are determined based on the diagnostic results. This invention obtains resource data from different data sources and parses the mapping relationship between network resource entities and different resource identifiers. This solves the problems of data silos and information inconsistencies caused by diverse data sources and inconsistent identifiers. Furthermore, it can associate scattered entity information based on the mapping relationship to construct an accurate and complete resource topology and form a unified resource profile. Intelligent diagnosis based on the unified resource profile can replace the traditional inefficient and error-prone manual verification, thereby accurately and automatically identifying problems such as missing data or incorrect topology relationships. This automates and intelligentizes the inventory of home and customer resources, effectively solving the problems of low efficiency, data inconsistency with reality, and unclear resource base caused by reliance on manual operations, and greatly improving the accuracy and real-time performance of resource inventory. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 This is one of the flowcharts of the intelligent inventory method for household customer resources provided by the present invention.

[0022] Figure 2 This is a schematic diagram of the process for constructing resource topology relationships provided by the present invention.

[0023] Figure 3 This is the second flowchart of the intelligent inventory method for household customer resources provided by the present invention.

[0024] Figure 4 This is a schematic diagram of the intelligent inventory device for household resources provided by the present invention.

[0025] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0027] The terms "first," "second," etc., used in this invention are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and that the objects distinguished by "first," "second," etc., are generally of the same class.

[0028] It should be noted that the actions of obtaining a user's personal signals, information or data mentioned in the various embodiments of the present invention are all carried out in compliance with the relevant data protection laws and policies of the country where the user is located, and with the authorization of the user and the owner of the corresponding device.

[0029] This invention provides a method for intelligent inventory of customer resources. Figure 1 This is one of the flowcharts illustrating the intelligent inventory method for household customer resources provided by this invention, such as... Figure 1 As shown, the method includes the following: Step 110: Obtain resource data of multiple network resource entities related to home customer business from different data sources; wherein, the same network resource entity has different resource identifiers in different data sources.

[0030] Specifically, firstly, resource data of multiple network resource entities related to home customer services are obtained from different data sources. The data source refers to a system or database that stores various types of information related to the operation of home customer (hereinafter referred to as "home customer") services. In a broader definition, any system capable of providing information such as network resource status, configuration, performance, and alarms can serve as the data source for this invention.

[0031] Here, the data source may include the operator's resource capability center system, network management system (network management system), alarm system, and external data system, etc., and the embodiments of the present invention do not specifically limit this.

[0032] The operator's resource capability center system typically stores basic ledger data on network resources, such as static or semi-static data on optical cables, splitters, ports, and user access information. The network management system monitors the operational status of network equipment, obtaining real-time performance data such as the device's central processing unit (CPU), port traffic, and optical power. The alarm system collects and manages alarm events occurring in the network, providing real-time or historical network alarm information. External data systems, such as geographic information systems (GIS), can provide geographical location information for network resources, assisting in resource location and on-site maintenance dispatch.

[0033] The methods for obtaining resource data from multiple network resource entities related to home customer services from different data sources are flexible and can be selected according to the characteristics of different data sources. For example, resource data can be actively pulled through application programming interfaces (APIs); periodic query tasks can be performed using direct database connections; data streams pushed in real time from data sources can be received through a publish-subscribe model; and in some special cases, semi-automatic methods such as file import can also be supported for obtaining resource data. In a preferred embodiment of the present invention, the acquisition process is automatically executed by an intelligent agent system to achieve unattended and routine data aggregation.

[0034] Here, the resource data of multiple network resource entities related to home customer services refers to the collection of information on all physical and logical resources involved in providing and ensuring network services to home customers. Resource data is characterized by being multi-source and heterogeneous. Specifically, resource data may include basic resource data, dynamic operational data, event-related data, and geographic location data, etc., but this embodiment of the invention does not impose specific limitations on these.

[0035] The basic resource data may include optical cable information, optical splitter information, optical junction box / fiber distribution box information, optical line terminal (OLT) equipment information, optical network unit (ONU) / optical network terminal (ONT) equipment information, port information, user account information, IP (Internet Protocol Address) information, etc., and the embodiments of the present invention do not specifically limit this.

[0036] Dynamic operational data can include performance metrics such as real-time / historical port traffic, optical power attenuation, and device CPU / memory utilization. Event-based data can include device online / offline alarms, port anomaly alarms, and optical path interruption alarms. Geographic location data can include the latitude and longitude coordinates of the device or optical cable route.

[0037] In this context, the same network resource entity may have different resource identifiers in different data sources.

[0038] Here, a network resource entity refers to a physical or logical unit in a communication network that can be independently identified and managed. Network resource entities are the basic elements that constitute network topology and service paths. For example, a user account, an ONU device, a physical port on an OLT device, and an optical splitter are all network resource entities.

[0039] A resource identifier is a string or value used to uniquely identify a network resource entity within a data source. The same network resource entity typically has different resource identifiers in different data sources. For example, in a resource capability center system, an ONU device might use an internal management ID, such as "ONU_001," as its resource identifier; while in a network management system, the same ONU device might use its Media Access Control Address (MAC), such as "XX-XX-XX-XX-XX-XX." This inconsistency is a major cause of data silos and difficulties in correlation analysis.

[0040] Furthermore, in addition to the resource data directly related to home customer services mentioned above, this embodiment of the invention can also obtain more macro-level global status information from the resource capability center monitoring system, providing richer diagnostic context. Specifically, the global status information monitored by the resource capability center includes at least the computing resource load, storage performance, network bandwidth, service health, and network quality data between data centers for each data center. Incorporating this global status information into the analysis helps determine whether the root cause of regional or widespread resource anomalies stems from the numerous end-user resources themselves or from performance bottlenecks or health issues in the data center infrastructure they depend on, thereby greatly improving the efficiency of locating the root cause of large-scale failures.

[0041] Step 120: Parse the resource data to obtain the mapping relationship between the network resource entity and the resource identifier.

[0042] Specifically, after obtaining the resource data, the resource data can be parsed to obtain the mapping relationship between network resource entities and resource identifiers.

[0043] Resource data parsing primarily involves data cleaning, standardization, and key information extraction to identify network resource entities and their corresponding resource identifiers from various data sources. For example, extracting information such as device name and port number from a single row of raw data containing multiple fields. The parsing process may also include data preprocessing, such as handling null values, outliers, and duplicate values, to improve data quality.

[0044] Here, the mapping relationship is used to reflect the correspondence between various resource identifiers of the same network resource entity in different data sources. The mapping relationship can be a mapping table, an association dictionary, or a set of association rules. For example, establishing a mapping relationship between "ONU_ID: ONU_001" and "MAC address: XX-XX-XX-XX-XX-XX".

[0045] Step 130: Based on the mapping relationship, construct the resource topology relationship between the network resource entities, so as to determine the resource profile of the resource data based on the resource topology relationship.

[0046] Specifically, after obtaining the mapping relationship, the resource topology relationship between network resource entities can be constructed based on the mapping relationship, so as to determine the resource profile of the resource data based on the resource topology relationship.

[0047] Here, resource topology relationships are used to reflect the connection structure and hierarchical relationships formed between network resource entities to carry services. The process of constructing resource topology relationships uses the mapping relationships obtained in previous steps as a unified basis, associating and integrating different identifiers representing the same physical entity. Then, according to the established business logic and physical connection rules, these identified entities are organized into a network topology or tree hierarchy according to the logical structure. For example, by identifying that "user account A" is bound to "ONU device B", and "ONU device B" is connected to "port 3 of optical splitter C" via a physical link, and "optical splitter C" further aggregates to "port 5 of OLT device D", a complete end-to-end service link topology extending from the user side to the core network side can be constructed.

[0048] Resource profiling is used to reflect a unified, multi-dimensional, and dynamic resource information view formed by integrating various static and dynamic attribute information of network resource entities, with resource topology relationships as the framework. Resource profiling is not merely a static connection diagram, but a panoramic view that dynamically reflects the resource status. Determining a resource profiling can be understood as attaching the parsed resource attribute information, such as device model, port status, and user service type, to the corresponding network resource entity nodes in the resource topology relationships, forming an information-rich topology structure.

[0049] Step 140: Obtain the diagnostic results of the intelligent diagnosis of the resource profile.

[0050] Specifically, after obtaining the resource profile, the diagnostic results of intelligent diagnosis of the resource profile can be obtained.

[0051] Here, intelligent diagnosis of resource profiles refers to using automated methods to verify and analyze resource profiles in order to discover problems such as inconsistent, incomplete, or unreasonable data.

[0052] The diagnostic result refers to the output of the intelligent diagnostic process, used to indicate any anomalies discovered. The diagnostic result can be a list of anomaly records, a set of alarm messages, or a structured diagnostic report. For example, the diagnostic result could be "User account user001 is in use, but its associated OLT port has been offline for more than 24 hours, indicating a data inconsistency anomaly"; or "ONU device 'ONU_001' is in an assigned state in the topology, but is not connected to any splitter, indicating an incomplete topology relationship."

[0053] Step 150: Based on the diagnostic results, determine the results of the resource data cleanup.

[0054] Specifically, after obtaining the diagnostic results, the results of the resource data inventory can be determined based on the diagnostic results.

[0055] The diagnostic results reflect the final state of the resource data after the inventory check and the necessary follow-up measures. Determining the inventory results may involve classifying and summarizing the diagnosed anomalies and clarifying their handling recommendations. For example, the inventory results may be a corrected resource dataset, a list of pending data issues, or a set of repair work orders to be dispatched. For instance, for diagnosed idle resources, the inventory results may mark them as awaiting recycling; for data missing issues, the inventory results may trigger a data entry process.

[0056] The method provided in this invention obtains resource data related to home customer business from different data sources, wherein the same network resource entity has different resource identifiers in different data sources; the resource data is parsed to obtain the mapping relationship between network resource entities and resource identifiers; based on the mapping relationship, a resource topology relationship between network resource entities is constructed, and a resource profile of the resource data is determined based on the resource topology relationship; intelligent diagnosis is performed on the resource profile to obtain a diagnosis result, and the investigation result of the resource data is determined based on the diagnosis result. This invention obtains resource data from different data sources and parses the mapping relationship between network resource entities and different resource identifiers. This solves the problems of data silos and information inconsistencies caused by diverse data sources and inconsistent identifiers. Furthermore, it can associate scattered entity information based on the mapping relationship to construct an accurate and complete resource topology and form a unified resource profile. Intelligent diagnosis based on the unified resource profile can replace the traditional inefficient and error-prone manual verification, thereby accurately and automatically identifying problems such as missing data or incorrect topology relationships. This automates and intelligentizes the inventory of home and customer resources, effectively solving the problems of low efficiency, data inconsistency with reality, and unclear resource base caused by reliance on manual operations, and greatly improving the accuracy and real-time performance of resource inventory.

[0057] Based on the above embodiments, step 130, which involves constructing the resource topology relationship between the network resource entities based on the mapping relationship, includes: Step 130-1: Obtain the user account, use the user account as the starting network resource entity, and create a link structure including the starting network resource entity; Step 130-2: Take the starting network resource entity as the current network resource entity, query the next-hop network resource entity of the current network resource entity in the mapping relationship, and add the next-hop network resource entity to the link structure; Step 130-3: Update the next-hop network resource entity to the current network resource entity until the current network resource entity is a preset core network port entity, and use the link structure as the resource topology relationship.

[0058] Specifically, Figure 2 This is a schematic diagram of the process for constructing resource topology relationships provided by the present invention, such as... Figure 2 As shown, firstly, the user account can be obtained, and the user account can be used as the starting network resource entity to create a link structure that includes the starting network resource entity.

[0059] In this context, a user account is a logical identifier used to uniquely identify a household customer's service transaction instance, such as a broadband internet account or a service account. The user account serves as the logical starting point for tracing resource paths from a business perspective.

[0060] In this context, the initial network resource entity is the first network resource entity node in the resource topology relationship construction process. In this embodiment, the user account, as a logical entity, is explicitly used as the starting point of the construction process, enabling the constructed topology relationship to be directly associated with specific customer services.

[0061] Here, the link structure is a data structure used to store network resource entities in sequence. The link structure can be a list, array, or linked list, etc. The embodiments of the present invention do not specifically limit it. The purpose of the link structure is to record the sequence of network resource entities discovered one by one along the service carrying path, starting from the initial network resource entity.

[0062] This step first determines a starting point for the investigation, namely a specific user account, and initializes an empty link structure, then stores the user account as the first element in the link structure.

[0063] Secondly, the starting network resource entity is used as the current network resource entity. The next-hop network resource entity of the current network resource entity is queried in the mapping relationship, and the next-hop network resource entity is added to the link structure.

[0064] Here, the current network resource entity refers to the network resource entity that is currently being processed and serves as the starting point for the query during the hop-by-hop link construction process. In the first iteration of the loop, the current network resource entity is the same as the starting network resource entity, namely the user account.

[0065] A next-hop network resource entity refers to the next network resource entity that is directly connected to the current network resource entity in the resource topology. For example, the next-hop entity for a user account might be the ONU / ONT device it is bound to; the next-hop entity for an ONU device might be the optical splitter port it is connected to.

[0066] The system uses the current network resource entity as an index to query within the mapping relationship. Because the mapping relationship unifies the identifiers of different data sources, the system can find the next-hop network resource entity directly associated with the current network resource entity across data sources. For example, by querying the mapping relationship using a user account, the system finds the MAC address of the ONU device associated with that user; the ONU device represented by this MAC address is the next-hop network resource entity. This newly discovered next-hop network resource entity is then appended to the end of the link structure.

[0067] Finally, the next-hop network resource entity can be updated to the current network resource entity until the current network resource entity is the preset core network port entity, thus obtaining the resource topology. This step is a cyclical iterative process with a termination condition. After adding the next-hop network resource entity to the link structure, the system updates it to the new current network resource entity and then repeats the second step, i.e., querying the next-hop network resource entity of this new current network resource entity.

[0068] In this context, the core network port entity is a predefined type of network resource entity that marks the endpoint of the access link for home users. In fiber optic access networks, this typically refers to the physical or logical port on the OLT device that connects to the core network. Using the core network port entity as a termination condition ensures that the constructed topology focuses on the complete access link from the user to the central office.

[0069] This cyclical process continues, and the link structure extends over time, for example: user account -> ONU device -> splitter port -> OLT port. The loop terminates when the type of the newly discovered network resource entity is determined to be a preset core network port entity. At this point, the ordered sequence of entities stored in the link structure constitutes a complete resource topology from the specified user to the core network access point. Repeating the above process for all user accounts constructs the resource topology of the entire home customer service network.

[0070] The method provided in this invention, by starting from the user account and iterating hop by hop to the core network port, can concretize the abstract topology construction process into a clear algorithm flow. This allows for the automated and deterministic tracing and generation of the complete service carrying link for each user using mapping relationships. This ensures the integrity and accuracy of the constructed resource topology relationship, effectively avoiding omissions or errors caused by manual sorting due to complex connection relationships, and improving the accuracy and reliability of subsequent resource profile construction.

[0071] Based on the above embodiments, step 130, which involves determining the resource profile of the resource data based on the resource topology relationship, includes: Step 1301: Generate a graph structure based on the resource topology relationship; wherein, the nodes in the graph structure correspond one-to-one with the network resource entities, and the edges in the graph structure represent the connection relationships of the network resource entities; Step 1302: Obtain dynamic operating parameters related to the network resource entity; the dynamic operating parameters are determined based on a performance attribute table or an alarm attribute table; Step 1303: Based on the resource identifier of the network resource entity, associate the dynamic operating parameters with the nodes in the graph structure to obtain the resource profile.

[0072] Specifically, firstly, a graph structure can be generated based on the resource topology, where each node in the graph structure corresponds one-to-one with a network resource entity.

[0073] Here, a graph structure is a mathematical structure used to represent objects (nodes) and their relationships (edges). In this invention, the graph structure transforms resource topology relationships into a standardized data model more suitable for complex analysis and computation. Nodes in the graph structure correspond one-to-one with various network resource entities in the resource topology relationships, such as user accounts, ONUs, and OLT ports, while edges in the graph structure represent the connections between these network resource entities.

[0074] Secondly, obtain the dynamic operating parameters related to the network resource entity, wherein the dynamic operating parameters are determined based on the performance attribute table or alarm attribute table.

[0075] Here, dynamic operating parameters refer to data that reflects the real-time or recent operational status of network resource entities and changes over time. Dynamic operating parameters are contrasted with the static configuration attributes of the entity, such as device model and port description. Dynamic operating parameters are a key basis for judging the health status and usage of resources.

[0076] The performance attribute table typically originates from the network management system and is a database table or data set storing performance metrics of storage devices or ports. For example, the performance attribute table may include fields such as timestamp, device ID, port ID, CPU utilization, memory utilization, uplink / downlink traffic, optical power transmission and reception values, and packet error rate. This embodiment of the invention does not impose specific limitations on these fields.

[0077] The alarm attribute table, typically derived from the alarm monitoring system, is a record table storing abnormal events occurring in the network. For example, the alarm attribute table may include fields such as alarm ID, alarm time, alarm object ID, alarm level, and alarm content; however, this embodiment of the invention does not impose specific limitations on these fields.

[0078] Finally, based on the resource identifiers of network resource entities, the dynamic operating parameters are associated with nodes in the graph structure to obtain a resource profile. Specifically, the system retrieves records from the performance attribute data table; for example, a resource identifier of MAC address AA:BB:CC:11:22:33 with a downlink traffic of 50Mbps. According to the pre-established mapping relationship, the system determines that this MAC address corresponds to node N_ONU_123 in the graph structure. Subsequently, the system associates and assigns the dynamic operating parameter of downlink traffic of 50Mbps as an attribute of node N_ONU_123.

[0079] By performing the aforementioned association operations on all available dynamic operating parameters, each node in the graph structure is endowed with a rich set of real-time changing attribute information. Ultimately, this comprehensive graph structure, integrating static configuration information and dynamic operating parameters, constitutes a resource profile. This resource profile is a digital twin model that comprehensively, dynamically, and accurately reflects the true state of network resources.

[0080] The method provided in this invention transforms resource topology relationships into a graph structure and associates them with dynamic operating parameters from external data sources such as performance and alarms. This enriches the original static topology representation, which only describes connection relationships, into a dynamic information model that integrates real-time operating status. The resource profile constructed in this way breaks through the limitations of traditional static descriptions, not only representing the basic attributes of resource entities but also dynamically reflecting their actual operating status. This effectively solves the technical deficiency of traditional resource inventory methods in being unable to perceive and present dynamic changes in resources in real time.

[0081] Based on the above embodiments, step 140 includes: Step 1401: Perform deterministic verification on the resource profile based on the rule engine to obtain deterministic anomaly records; Step 1402: Extract the portrait features of the resource portrait, input the portrait features into the diagnostic model, and obtain the probabilistic anomaly alarm output by the diagnostic model; the diagnostic model is trained based on the sample resource portrait and the label probabilistic anomaly alarm corresponding to the sample resource portrait. Step 1403: Determine the diagnostic result based on the deterministic abnormal records and / or the probabilistic abnormal alarms.

[0082] Specifically, firstly, a deterministic check can be performed on the resource profile based on the rule engine to obtain deterministic anomaly records.

[0083] The rules engine is a software component capable of executing predefined business rules. These rules are typically expressed in "IF-THEN" format, with clear logic and explicit judgment criteria. Deterministic verification refers to checking resource profiles based on these hard-coded rules. For example, a rule can be defined as: "IF an ONU entity has been assigned to a user AND the ONU entity does not have an uplink splitter port in the resource topology THEN record as a topology incomplete anomaly."

[0084] Here, the deterministic anomaly log is the output of the rule engine after performing deterministic checks; it is a list of anomaly events that are 100% confirmed to have violated established rules. This step is mainly used to discover errors that have clear judgment criteria and are logically or configurationally obvious. Examples include integrity checks (such as empty required fields), consistency checks (such as inconsistent states of the same device in two systems), and logical checks. Logical checks can include issues such as VLAN (Virtual Local Area Network) configuration mismatches with service types.

[0085] Then, the profile features of the resource profile can be extracted and input into the diagnostic model to obtain probabilistic anomaly alerts output by the diagnostic model. This step uses machine learning methods to discover potential anomalies that are difficult to describe by rules and are hidden in data patterns.

[0086] Among them, profile features are used to reflect the characteristic information of resource profiles. For example, for a port node, its average traffic over the past 7 days, the number of alarms in the past 24 hours, the total number of ports of its optical splitter, and the percentage of used ports of its optical splitter can be extracted and combined into a profile feature.

[0087] The diagnostic model is a pre-trained artificial intelligence model. The training steps for the diagnostic model include: First, collect sample resource profiles and corresponding label probabilistic anomaly alarms. Label probabilistic anomaly alarms can come from fault work orders confirmed after the fact, anomalies manually annotated by experts, etc.

[0088] Then, the sample profile features corresponding to the sample resource profile can be input into the initial diagnostic model to obtain the predicted probabilistic anomaly alarm output by the initial diagnostic model. The initial diagnostic model can be an XGBoost model or a LightGBM model, which learns anomaly classification patterns when labeled, enabling it not only to determine whether an anomaly is present but also to identify the type of anomaly, such as suspected unauthorized connection or port aging.

[0089] Here, the predictive probabilistic anomaly alert is the output of the diagnostic model after inferring the features of a newly input sample profile. A predictive probabilistic anomaly alert is typically a probability value or an anomaly score, representing the likelihood that the network resource entity belongs to a certain type of anomaly.

[0090] After obtaining the predicted probabilistic anomaly alarm based on the initial diagnostic model, the prediction loss can be determined based on the difference between the predicted probabilistic anomaly alarm and the labeled probabilistic anomaly alarm. The initial diagnostic model can then be trained based on the prediction loss, and the trained initial diagnostic model can be used as the diagnostic model.

[0091] Finally, a diagnostic result can be determined based on deterministic anomaly records and / or probabilistic anomaly alerts. The diagnostic result is a comprehensive conclusion, which may include deterministic problems detected by the rule engine and probabilistic problems detected by the diagnostic model. It should be understood that a resource entity may only trigger a rule (deterministic anomaly), may only be flagged by the diagnostic model (probabilistic anomaly), or may be detected by both. These results are combined to obtain the diagnostic result.

[0092] The method provided in this invention adopts a dual-track diagnostic scheme that combines a rule engine and a diagnostic model. This allows the deterministic logical judgment and probabilistic pattern discovery capabilities to complement each other. As a result, it can not only efficiently and accurately detect explicit errors such as data integrity and consistency, but also intelligently identify potential anomalies and fault risks hidden in complex data patterns that are difficult to define by traditional rules. This can greatly improve the depth and breadth of intelligent diagnosis of customer resources.

[0093] Based on the above embodiments, step 1402, which involves extracting the image features of the resource image, includes: Step 1402-1: Obtain the statistical value of the historical port traffic of the network resource entity within a preset period; Step 1402-2: Determine the alarm frequency of the network resource entity within the preset period; Step 1402-3: Obtain the operating status data of other network resource entities that belong to the same parent network resource entity as the network resource entity; Step 1402-4: Determine the profile features based on the statistical values, the alarm frequency, and the operating status data.

[0094] Specifically, firstly, statistical values ​​of the historical port traffic of network resource entities within a preset period can be obtained. The preset period refers to a predefined time window, such as the last 7 days, 30 days, or 90 days, etc., and this embodiment of the invention does not specifically limit this. The preset period is set to capture the behavioral trend of resource entities over a period of time, rather than their instantaneous state.

[0095] The statistical value of historical port traffic refers to a statistical indicator obtained by calculating port traffic data collected within a preset period. This does not simply refer to the raw traffic data points, but rather to the processed and refined characteristics.

[0096] Here, the statistical values ​​of port historical traffic may include the mean and variance of port historical traffic, the maximum and minimum values ​​of port historical traffic, the quantiles of port historical traffic, etc., and the embodiments of the present invention do not specifically limit them.

[0097] The mean and variance are used to determine the average level and volatility of historical port traffic. The maximum and minimum values ​​are used to capture the peaks and troughs of historical port traffic. The quantile, which can be the 95th percentile of historical port traffic, is used to assess the typical high load level of the port.

[0098] Then, the alarm frequency of a network resource entity within a preset period can be determined. Alarm frequency refers to the number of times a network resource entity triggers an alarm event within the preset period. High-frequency alarms, even low-level alarms, may indicate that the entity is in an unstable state or is about to fail; for example, port aging may cause ports to frequently go online and offline.

[0099] Furthermore, it is possible to obtain the operational status data of other network resource entities that belong to the same parent network resource entity as the network resource entity. Here, the same parent network resource entity refers to a sibling entity that shares the same parent node with the currently analyzed network resource entity in the resource topology. For example, for a port under a splitter, the other network resource entities under the same parent network resource entity refer to all other ports under that splitter.

[0100] Here, the running status data refers to the current working status of sibling entities that share the same parent node with the network resource entity being analyzed, such as online, offline, idle, occupied, etc. This embodiment of the invention does not specifically limit this.

[0101] Obtaining the operational status data of other network resource entities that belong to the same parent network resource entity as the network resource entity is to provide contextual information. For example, if all ports under a splitter are offline, the problem is likely with the splitter or its uplink optical path; however, if only one port is offline while all other ports are normal, the problem is more likely limited to that port itself.

[0102] Finally, profile features can be determined based on statistical values, alarm frequency, and operational status data. This step combines the extracted multi-dimensional information to form a numerical feature vector, i.e., profile features.

[0103] The method provided in this invention determines profile features based on statistical values, alarm frequency, and operating status data, thereby improving the accuracy and reliability of profile features.

[0104] Based on the above embodiments, step 1401 includes: Step 1401-1: Verify the consistency between the running status data of the network resource entity in the resource profile and the status data of the associated port of the network resource entity to obtain the first record; Step 1401-2: Perform a logical check on the configuration data related to network services in the resource profile to obtain the second record; Step 1401-3: Based on the first record and / or the second record, determine the deterministic anomalous record.

[0105] Specifically, firstly, the consistency between the operational status data of network resource entities in the resource profile and the status data of their associated ports can be verified to obtain the first record. This step focuses on the mutual verification of data across different dimensions. For example, a user's service status and the status of the physical port carrying that service should match. The first record is a list specifically documenting data consistency issues.

[0106] A typical verification scenario is as follows: the rule engine detects that a user account's network resource entity is in use, but the associated OLT port entity in the resource profile is offline for more than 24 hours. These two states are clearly contradictory and constitute data inconsistency. The rule engine will record this situation, creating a primary record.

[0107] Next, the configuration data related to network services in the resource profile can be logically checked to obtain the second record. This step mainly checks whether the resource configuration conforms to the basic logic of service activation and network operation. The second record is a list specifically recording logical configuration errors.

[0108] A typical verification scenario is that the rule engine discovers that the VLAN ID configured for a user does not match the VLAN planning required for their subscribed service type, such as high-speed internet or IPTV, or that the VLAN ID is not actually configured on the corresponding OLT device. While this configuration error is not considered a state inconsistency, it directly leads to service disruption and is a logical error. The rule engine will record this type of problem, creating a secondary record.

[0109] Finally, deterministic anomaly records can be obtained based on the first record and / or the second record.

[0110] Based on the above embodiments, step 150 includes: Step 150-1: Based on the diagnostic results, determine the treatment type; Step 150-2: When the handling type is automated repair, trigger the data repair process to obtain the investigation results; Step 150-3: When the handling type is manual handling, a maintenance work order is generated, and the investigation result is determined based on the maintenance result of the maintenance work order.

[0111] Specifically, firstly, based on the diagnostic results, the treatment type is determined. The treatment types are mainly divided into two categories: automated repair and manual treatment.

[0112] Here, the basis for determining the treatment type may include the type and complexity of the anomaly, the risk level of the repair operation, and the confidence level of the diagnostic results, etc., and the embodiments of the present invention do not specifically limit this.

[0113] Then, when the handling type is automated repair, a data repair process is triggered to obtain the investigation results. For diagnostic records determined to be automatically repaired, the system will automatically execute predefined repair scripts or call relevant system interfaces to correct the data. After successful repair, the updated data status becomes part of the final investigation results, and detailed automated operation logs are recorded. A rollback mechanism can even be designed to ensure security.

[0114] When the handling type is manual, a maintenance work order is generated. Based on the maintenance results of the work order, the investigation results are determined. A maintenance work order is an electronic task instruction sheet used to assign problems requiring manual handling to specific maintenance personnel.

[0115] For diagnostic records determined to require manual intervention, the system automatically creates a maintenance work order. This work order details the identified problem (diagnostic result), the involved network resource entities, their geographical locations (if available), and recommended verification actions. The work order is then dispatched to maintenance personnel. After completing on-site verification or remote handling, the maintenance personnel will report the maintenance results to the system, such as confirming line damage and replacement, or confirming unauthorized user connections and removal. This final result, verified and processed manually, constitutes another part of the investigation results.

[0116] The method provided by this invention introduces a treatment type judgment based on diagnostic results and establishes two closed-loop processes: automated repair and manual handling. This enables differentiated and intelligent management of problems discovered during the investigation. It allows a large number of simple and clear errors to be quickly processed through automated processes, while complex problems requiring on-site confirmation are structured and transferred to manual handling. This forms a complete management closed loop from problem discovery to problem resolution, greatly improving the overall efficiency of resource investigation work.

[0117] Based on the above embodiments, step 150-3, which involves generating a maintenance work order when the handling type is manual handling, includes: Step 210: When the handling type is manual handling, obtain the tasks to be processed related to the diagnosis result, as well as the candidate information of each maintenance personnel; the candidate information includes the skill tags, real-time load and geographical location of each maintenance personnel. Step 220: Based on the complexity of the task to be processed, as well as the skill tags, real-time load and geographical location of each maintenance personnel, determine the target maintenance personnel; Step 230: Dispatch a maintenance work order, including the task to be processed, to the target maintenance personnel.

[0118] Specifically, first, the pending tasks related to the diagnostic results are obtained, along with the candidate information for each maintenance personnel. The pending tasks are the specific work content derived from the diagnostic results that requires manual completion.

[0119] Here, the candidate information refers to the dynamic set of attributes of all maintenance personnel in the system who can receive work orders. This candidate information includes each maintenance personnel's skill tags, real-time load, and geographical location.

[0120] The skill tags for each maintenance worker are keywords describing their professional capabilities, such as fiber optic fusion splicing or OLT equipment expert. Real-time workload indicates the current workload of a maintenance worker, such as the number of pending work orders and estimated total working hours. Geographic location is the real-time coordinates of the maintenance worker, used to calculate their physical distance from the location of the pending tasks.

[0121] Then, the target maintenance personnel can be identified based on the complexity of the task to be processed, as well as the skill tags, real-time load, and geographical location of each maintenance personnel.

[0122] The complexity of the task to be processed is a quantitative assessment of the task's difficulty. For example, a simple ONU status check has low complexity, while a survey task involving trunk optical cables has high complexity. The complexity of the task to be processed can be associated with the required skill tags.

[0123] In one specific embodiment, firstly, maintenance personnel with the required skill tags are selected based on the complexity and type of the task to be processed. Among the skill-matching maintenance personnel, those with lower real-time loads are given priority to avoid excessive task concentration. Based on the first two steps of selection, personnel geographically closest to the task location are chosen to shorten travel time and improve response speed. The target maintenance personnel are determined by the algorithm to be the most suitable personnel to perform the task after the above multi-dimensional comprehensive evaluation.

[0124] Finally, a maintenance work order containing the pending tasks is dispatched to the target maintenance personnel.

[0125] The method provided in this invention introduces an intelligent dispatch algorithm that incorporates skill tags, real-time load, and geographical location of each maintenance worker when generating maintenance work orders. This transforms traditional random dispatch or manual scheduling into data-driven precise matching, ensuring that suitable tasks are assigned to the target maintenance personnel with the most relevant skills, the most timeliness, and the lowest cost. This optimizes the allocation of human resources, significantly improves the efficiency and quality of manual handling, effectively reduces the average fault handling time, and increases the first-time repair success rate.

[0126] Based on any of the above embodiments Figure 3 This is the second flowchart of the intelligent inventory method for household customer resources provided by the present invention, as shown below. Figure 3 As shown, the system first performs data access and fusion steps, acquiring resource data related to home customer services from different data sources. The resource data is then parsed to obtain the mapping relationship between network resource entities and resource identifiers. Based on this mapping relationship, a resource topology relationship is constructed between network resource entities, and based on this topology relationship, a resource profile is determined for the resource data. After obtaining a unified data foundation, the system proceeds to the intelligent verification and diagnosis step, using a rule engine and diagnostic model to conduct in-depth analysis of the resource profile to detect anomalies.

[0127] The diagnosed problem will be classified. If it is determined to be a simple data error, such as a clear data inconsistency, the process will enter the automatic repair branch, where the system will automatically perform data correction. If it is determined to be a complex on-site problem, such as a potential fault requiring manual inspection, the process will enter the work order generation and dispatch branch, assigning the task to the most suitable personnel.

[0128] Whether a work order is automatically repaired or manually processed, its execution results are incorporated into the result feedback and verification step. In this step, the system verifies the effectiveness of the repair or processing. If verification fails, it means the problem has not been properly resolved or new problems have been introduced, and the process returns to the intelligent verification and diagnosis step, forming a closed loop of error correction and re-analysis. If verification succeeds, it indicates the problem has been resolved, and the process continues to the resource profile update step to ensure the resource data model remains synchronized with the real world. Finally, the latest resource status after the investigation and repair can be presented to managers through visualization, completing the closed-loop process of the entire intelligent investigation.

[0129] The intelligent inventory device for household and customer resources provided by the present invention is described below. The intelligent inventory device for household and customer resources described below can be referred to in correspondence with the intelligent inventory method for household and customer resources described above.

[0130] Based on any of the above embodiments, the present invention provides a smart inventory device for residential customer resources. Figure 4This is a schematic diagram of the intelligent inventory device for household resources provided by the present invention, as shown below. Figure 4 As shown, the device includes: The acquisition unit 410 is used to acquire resource data of multiple network resource entities related to home customer business from different data sources; wherein the same network resource entity has different resource identifiers in different data sources. The parsing unit 420 is used to parse the resource data to obtain the mapping relationship between the network resource entity and the resource identifier; The construction unit 430 is used to construct the resource topology relationship between the network resource entities based on the mapping relationship, so as to determine the resource profile of the resource data based on the resource topology relationship; The diagnostic unit 440 is used to obtain the diagnostic results of intelligent diagnosis of the resource profile; The determining unit 450 is used to determine the results of the resource data investigation based on the diagnostic results.

[0131] This invention provides an apparatus for acquiring resource data related to home customer services from different data sources, wherein the same network resource entity has different resource identifiers in different data sources; parsing the resource data to obtain the mapping relationship between network resource entities and resource identifiers; constructing resource topology relationships between network resource entities based on the mapping relationship; determining resource profiles of resource data based on the resource topology relationships; performing intelligent diagnosis on the resource profiles to obtain diagnostic results; and determining the investigation results of resource data based on the diagnostic results. This invention obtains resource data from different data sources and parses the mapping relationship between network resource entities and different resource identifiers. This solves the problems of data silos and information inconsistencies caused by diverse data sources and inconsistent identifiers. Furthermore, it can associate scattered entity information based on the mapping relationship to construct an accurate and complete resource topology and form a unified resource profile. Intelligent diagnosis based on the unified resource profile can replace the traditional inefficient and error-prone manual verification, thereby accurately and automatically identifying problems such as missing data or incorrect topology relationships. This automates and intelligentizes the inventory of home and customer resources, effectively solving the problems of low efficiency, data inconsistency with reality, and unclear resource base caused by reliance on manual operations, and greatly improving the accuracy and real-time performance of resource inventory.

[0132] Based on any of the above embodiments, the construction unit 430 is specifically used for: Obtain a user account, use the user account as the starting network resource entity, and create a link structure including the starting network resource entity; The starting network resource entity is used as the current network resource entity. The next-hop network resource entity of the current network resource entity is queried in the mapping relationship, and the next-hop network resource entity is added to the link structure. The next-hop network resource entity is updated to the current network resource entity until the current network resource entity is a preset core network port entity, and the link structure is used as the resource topology relationship.

[0133] Based on any of the above embodiments, the construction unit 430 is specifically used for: A graph structure is generated based on the resource topology; wherein, the nodes in the graph structure correspond one-to-one with the network resource entities, and the edges in the graph structure represent the connection relationships of the network resource entities; Obtain dynamic operating parameters related to the network resource entity; the dynamic operating parameters are determined based on a performance attribute table or an alarm attribute table. Based on the resource identifier of the network resource entity, the dynamic operating parameters are associated with the nodes in the graph structure to obtain the resource profile.

[0134] Based on any of the above embodiments, the diagnostic unit 440 specifically includes: The verification unit is used to perform deterministic verification on the resource profile based on the rule engine to obtain deterministic anomaly records; An extraction unit is used to extract the portrait features of the resource portrait, input the portrait features into the diagnostic model, and obtain the probabilistic anomaly alarm output by the diagnostic model; the diagnostic model is trained based on the sample resource portrait and the label probabilistic anomaly alarm corresponding to the sample resource portrait. A diagnostic result determination unit is used to determine the diagnostic result based on the deterministic abnormality record and / or the probabilistic abnormality alarm.

[0135] Based on any of the above embodiments, the extraction unit is specifically used for: Obtain statistical values ​​of the historical port traffic of the network resource entity within a preset period; Determine the alarm frequency of the network resource entity within the preset period; Obtain the operational status data of other network resource entities that belong to the same parent network resource entity as the network resource entity; Based on the statistical values, the alarm frequency, and the operating status data, the profile features are determined.

[0136] Based on any of the above embodiments, the verification unit is specifically used for: The consistency between the running status data of the network resource entity in the resource profile and the status data of the associated port of the network resource entity is checked to obtain the first record; Logical verification is performed on the configuration data related to network services in the resource profile to obtain a second record; Based on the first record and / or the second record, the deterministic anomalous record is determined.

[0137] Based on any of the above embodiments, the determining unit 450 specifically includes: A type determination unit is used to determine the treatment type based on the diagnostic results; The first processing unit is used to trigger a data repair process to obtain the investigation results when the disposal type is automated repair; The second processing unit is used to generate a maintenance work order when the disposal type is manual disposal, and to determine the investigation result based on the maintenance result of the maintenance work order.

[0138] Based on any of the above embodiments, the second processing unit is specifically used for: When the handling type is manual handling, the pending tasks related to the diagnostic results are obtained, as well as the candidate information of each maintenance personnel; the candidate information includes the skill tags, real-time load and geographical location of each maintenance personnel. Based on the complexity of the task to be processed, as well as the skill tags, real-time load, and geographical location of each maintenance personnel, the target maintenance personnel are determined. A maintenance work order, including the task to be processed, is dispatched to the target maintenance personnel.

[0139] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5As shown, the electronic device may include: a processor 510, a communications interface 520, a memory 530, and a communication bus 540, wherein the processor 510, communications interface 520, and memory 530 communicate with each other through the communication bus 540. The processor 510 can call logical instructions in the memory 530 to execute a smart inventory method for home customer resources. This method includes: obtaining resource data of multiple network resource entities related to home customer business from different data sources; wherein the same network resource entity has different resource identifiers in different data sources; parsing the resource data to obtain a mapping relationship between the network resource entities and the resource identifiers; constructing a resource topology relationship between the network resource entities based on the mapping relationship to determine a resource profile of the resource data based on the resource topology relationship; obtaining a diagnostic result of intelligent diagnosis of the resource profile; and determining the inventory result of the resource data based on the diagnostic result.

[0140] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a 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.

[0141] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the intelligent inventory method for home customer resources provided by the above methods. The method includes: obtaining resource data of multiple network resource entities related to home customer business from different data sources; wherein the same network resource entity has different resource identifiers in different data sources; parsing the resource data to obtain a mapping relationship between the network resource entity and the resource identifier; constructing a resource topology relationship between the network resource entities based on the mapping relationship, so as to determine the resource profile of the resource data based on the resource topology relationship; obtaining a diagnostic result of intelligent diagnosis of the resource profile; and determining the inventory result of the resource data based on the diagnostic result.

[0142] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the intelligent inventory method for household customer resources provided by the methods described above. The method includes: obtaining resource data of multiple network resource entities related to household customer business from different data sources; wherein the same network resource entity has different resource identifiers in different data sources; parsing the resource data to obtain a mapping relationship between the network resource entities and the resource identifiers; constructing a resource topology relationship between the network resource entities based on the mapping relationship, so as to determine a resource profile of the resource data based on the resource topology relationship; obtaining a diagnostic result of intelligent diagnosis of the resource profile; and determining the inventory result of the resource data based on the diagnostic result.

[0143] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0144] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0145] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these 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 intelligent inventory of customer resources, characterized in that, include: Resource data of multiple network resource entities related to home customer business are obtained from different data sources; wherein the same network resource entity has different resource identifiers in different data sources; The resource data is parsed to obtain the mapping relationship between the network resource entity and the resource identifier; Based on the mapping relationship, a resource topology relationship between the network resource entities is constructed to determine the resource profile of the resource data based on the resource topology relationship. Obtain the diagnostic results of the intelligent diagnosis of the resource profile; Based on the diagnostic results, the results of the resource data clearance are determined.

2. The intelligent inventory method for residential customer resources according to claim 1, characterized in that, The step of constructing the resource topology relationship between the network resource entities based on the mapping relationship includes: Obtain a user account, use the user account as the starting network resource entity, and create a link structure including the starting network resource entity; The starting network resource entity is used as the current network resource entity. The next-hop network resource entity of the current network resource entity is queried in the mapping relationship, and the next-hop network resource entity is added to the link structure. The next-hop network resource entity is updated to the current network resource entity until the current network resource entity is a preset core network port entity, and the link structure is used as the resource topology relationship.

3. The intelligent inventory method for residential customer resources according to claim 1, characterized in that, The process of determining the resource profile of the resource data based on the resource topology includes: A graph structure is generated based on the resource topology; wherein, the nodes in the graph structure correspond one-to-one with the network resource entities, and the edges in the graph structure represent the connection relationships of the network resource entities; Obtain dynamic operating parameters related to the network resource entity; the dynamic operating parameters are determined based on a performance attribute table or an alarm attribute table. Based on the resource identifier of the network resource entity, the dynamic operating parameters are associated with the nodes in the graph structure to obtain the resource profile.

4. The intelligent inventory method for guest resources according to any one of claims 1 to 3, characterized in that, The step of obtaining the diagnostic results of the intelligent diagnosis of the resource profile includes: Based on the rule engine, a deterministic check is performed on the resource profile to obtain deterministic anomaly records; Extract the profile features of the resource profile, input the profile features into the diagnostic model, and obtain the probabilistic anomaly alarm output by the diagnostic model; the diagnostic model is trained based on the sample resource profile and the label probabilistic anomaly alarm corresponding to the sample resource profile. The diagnostic result is determined based on the deterministic anomaly record and / or the probabilistic anomaly alarm.

5. The intelligent inventory method for residential customer resources according to claim 4, characterized in that, The extraction of the image features of the resource image includes: Obtain statistical values ​​of the historical port traffic of the network resource entity within a preset period; Determine the alarm frequency of the network resource entity within the preset period; Obtain the operating status data of other network resource entities that belong to the same parent network resource entity as the network resource entity; Based on the statistical values, the alarm frequency, and the operating status data, the profile features are determined.

6. The intelligent inventory method for guest resources according to claim 4, characterized in that, The deterministic verification of the resource profile based on the rule engine to obtain deterministic anomaly records includes: The consistency between the running status data of the network resource entity in the resource profile and the status data of the associated port of the network resource entity is checked to obtain the first record; Logical checks are performed on the configuration data related to network services in the resource profile to obtain a second record; Based on the first record and / or the second record, the deterministic anomalous record is determined.

7. The intelligent inventory method for guest resources according to any one of claims 1 to 3, characterized in that, The determination of the inventory results of the resource data based on the diagnostic results includes: Based on the diagnostic results, the treatment type will be determined; When the handling type is automated repair, the data repair process is triggered to obtain the investigation results; When the handling type is manual handling, a maintenance work order is generated, and the investigation result is determined based on the maintenance result of the maintenance work order.

8. The intelligent inventory method for residential customer resources according to claim 7, characterized in that, When the handling type is manual handling, a maintenance work order is generated, including: When the handling type is manual handling, the pending tasks related to the diagnostic results are obtained, as well as the candidate information of each maintenance personnel; the candidate information includes the skill tags, real-time load and geographical location of each maintenance personnel. Based on the complexity of the task to be processed, as well as the skill tags, real-time load, and geographical location of each maintenance personnel, the target maintenance personnel are determined. A maintenance work order, including the task to be processed, is dispatched to the target maintenance personnel.

9. A smart inventory device for household customer resources, characterized in that, include: The acquisition unit is used to acquire resource data of multiple network resource entities related to home customer business from different data sources; wherein the same network resource entity has different resource identifiers in different data sources; The parsing unit is used to parse the resource data to obtain the mapping relationship between the network resource entity and the resource identifier; The construction unit is used to construct the resource topology relationship between the network resource entities based on the mapping relationship, so as to determine the resource profile of the resource data based on the resource topology relationship; A diagnostic unit is used to obtain the diagnostic results of intelligent diagnosis of the resource profile; The determining unit is used to determine the results of the inventory of the resource data based on the diagnostic results.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the intelligent inventory method for household resources as described in any one of claims 1 to 8.

Images