Inference method, system and apparatus

Abstract

Provided in the embodiments of the present application are an inference method, system and apparatus, which can be used by a cognitive inference service producer to generate inference solutions having confidence levels and / or risk values, so as to feed back an inference solution to a cognitive inference service consumer on the basis of the confidence levels and / or the risk values of the inference solutions, such that the cognitive inference service consumer can obtain a required inference solution. The method comprises: a cognitive inference service producer receiving an inference request, wherein the inference request is used for requesting the provision of an inference solution for achieving a user intent objective; on the basis of the inference request, acquiring a first inference solution set comprising at least one inference solution, and at least one of a confidence level and a risk value which correspond to each inference solution, wherein the confidence level is used for indicating the probability of achieving an inference task when an inference solution is used, and the risk value is used for indicating the probability of changing the user experience when the inference solution is used to achieve the inference task; and feeding back at least one solution in the first inference solution set.

Description

Reasoning method, system and device

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on December 7, 2023, with application number 202311683927.5 and application name “A Reasoning Method, System and Device”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communications, and in particular to an inference method, system, and device. Background Art

[0003] In communication systems, for example, in scenarios such as network optimization or maintenance, cognitive reasoning service consumers can request reasoning solutions from task reasoning service producers. These solutions, for example, are designed to achieve the desired optimization or maintenance goals. The task reasoning service producer generates the solution and then sends it to the cognitive reasoning service consumer. However, due to the complexity of network system topology and the interdependence of fault propagation, multiple root cause reasoning solutions may be possible. Therefore, ensuring that cognitive reasoning service consumers receive the required solution becomes a pressing issue.

[0004] Summary of the Invention

[0005] The embodiments of the present application provide a reasoning method, system, and device that can be used by cognitive reasoning service producers to generate reasoning solutions with confidence and / or risk values, thereby feeding back reasoning solutions to cognitive reasoning service consumers based on the confidence and / or risk values ​​of the reasoning solutions, so that cognitive reasoning service consumers can obtain the required reasoning solutions.

[0006] In view of this, in a first aspect, the present application provides an inference method, comprising:

[0007] An inference request is received, where the inference request is used to request a solution for achieving a reasoning task, where the reasoning task includes a task for completing a user's intended goal; a first inference solution set is then obtained according to the inference request, where the first inference solution set includes at least one inference solution and at least one of a confidence level or a risk value corresponding to each inference solution, where each inference solution is used to achieve the reasoning task, where the confidence level indicates the probability of achieving the reasoning task when the inference solution is used, and where the risk value indicates the probability of changing the user experience when the inference solution is used to achieve the reasoning task; and a second inference solution set is sent, where the second inference solution set includes at least one solution in the first inference solution set.

[0008] In the implementation mode of the present application, while generating the reasoning scheme, the confidence and / or risk value of each reasoning scheme is also generated. Therefore, when feeding back the reasoning scheme, the confidence and / or risk value of each reasoning scheme can be fed back at the same time, or the reasoning scheme can be screened based on the confidence of each reasoning scheme and then fed back, so that consumers can more accurately select the required scheme based on the received reasoning scheme.

[0009] For ease of understanding, the sender of the inference request is referred to as the consumer, and the receiver of the inference request is referred to as the producer. No further details will be given below.

[0010] In one possible implementation, the aforementioned sending of the second set of inference solutions may include: using the first set of inference solutions as the second set of inference solutions; and sending the second set of inference solutions. Therefore, in the implementation of the present application, the confidence level and / or risk value corresponding to each inference solution can be fed back simultaneously, so that consumers can accurately select the desired inference solution based on the confidence level and / or risk value of each inference solution based on the data they receive.

[0011] In one possible implementation, the aforementioned inference request also carries a confidence indicator and / or a risk indicator. The confidence indicator is used to request the provision of a confidence level corresponding to the inference solution, and the risk indicator is used to request the provision of a risk value corresponding to the inference solution. Therefore, the consumer can carry the confidence indicator and / or risk indicator in the inference request, allowing the producer to provide feedback on the confidence level and / or risk value of each inference solution based on the inference request, thereby allowing the consumer to accurately select the desired inference solution based on the confidence level and / or risk value of each inference solution based on the data received.

[0012] In a possible implementation, the aforementioned sending of the second reasoning solution set may include: screening out solutions with confidence levels higher than a confidence threshold and / or solutions with risk values ​​higher than a risk threshold from the first reasoning solution set to obtain a second reasoning solution set; and sending the second reasoning solution set.

[0013] Therefore, in the implementation manner of the present application, the producer can screen the generated reasoning solutions before feeding back the reasoning solutions to the consumers, thereby directly feeding back solutions that meet the needs of the consumers, so that the consumers can obtain the required reasoning solutions more accurately.

[0014] In one possible implementation, the inference request also carries a confidence threshold and / or a risk threshold. Therefore, in this implementation, the consumer can determine the confidence threshold and / or the risk threshold, allowing the producer to directly provide the consumer with an inference solution that meets their needs.

[0015] In one possible implementation, the aforementioned sending of the second set of inference solutions may further include: selecting one of the inference solutions from the first set of inference solutions as a solution for the second set of inference solutions based on at least one of the confidence level or risk value corresponding to each inference solution; and sending the second set of inference solutions. In this implementation, the producer can directly provide a single inference solution to the consumer, allowing the consumer to accurately obtain the desired inference solution without having to make a selection.

[0016] In one possible implementation, the aforementioned screening out one of the inference schemes from the first set of inference schemes according to at least one of the confidence or risk values ​​corresponding to each inference scheme may include: screening out one of the inference schemes that meets the screening conditions from the first set of inference schemes, where the screening conditions include at least one of the following: an inference scheme with the highest confidence, an inference scheme with the lowest risk value, or an inference scheme with the highest fusion value after fusing the confidence and risk values. In the implementation of the present application, an inference scheme with the highest confidence, the lowest risk value, or a fusion of the confidence and risk values ​​may be selected, so as to adapt to a variety of scenarios.

[0017] In one possible implementation, the inference request also carries a policy indication, which is used to indicate at least one of the screening conditions. Therefore, in the implementation of the present application, the consumer can select the screening condition, so that the producer can directly select the inference solution that meets the consumer's needs.

[0018] In one possible implementation, the second set of inference solutions also includes at least one of a confidence level or a risk value corresponding to the inference solution. Therefore, the producer can also provide feedback on the confidence level and / or risk value to the consumer, allowing the consumer to learn the confidence level and / or risk value corresponding to each inference solution. When multiple inference solutions are provided, the consumer can further select the desired solution based on the confidence level and / or risk value of each inference solution.

[0019] In one possible implementation, the aforementioned acquisition of the first set of inference solutions based on the inference request may include: obtaining monitoring data and a knowledge base corresponding to the inference task, where the monitoring data includes data on devices related to the inference task, and the knowledge in the knowledge base includes rules for completing the inference task; and obtaining the first set of inference solutions based on the monitoring data and knowledge. Therefore, in the implementation of the present application, inference can be performed based on locally stored knowledge and collected real-time monitoring data, thereby obtaining an accurate inference solution based on pre-set rules.

[0020] In one possible implementation, the aforementioned method may further include: obtaining an execution result, where the execution result is obtained after executing a reasoning scheme included in the second reasoning scheme set; and updating knowledge based on the execution result to obtain updated knowledge. Therefore, in the implementation of the present application, knowledge can be updated based on the execution result of the reasoning scheme, thereby making the knowledge stored in the knowledge base more accurate and improving the accuracy of subsequently generated reasoning schemes.

[0021] In one possible implementation, the aforementioned reasoning task may specifically include at least one of a fault location task, a fault repair task, a configuration verification task, a hidden danger identification task, or a solution recommendation task. The fault location task indicates locating the root cause of a device fault, the fault repair task indicates repairing the device that caused the fault, the configuration verification task indicates verifying the device configuration, the hidden danger identification task indicates obtaining the probability of a device fault, and the solution recommendation task indicates obtaining at least one solution for executing the target task. Therefore, the method provided in this application can be applied to a variety of network optimization or maintenance scenarios and has strong generalization capabilities.

[0022] In the second aspect, the present application provides a reasoning method, including: first, sending an inference request, the inference request is used to request the provision of a solution for achieving a reasoning task, and the reasoning task includes a task to complete the user's intended goal; then, receiving a second inference solution set, the second inference solution set includes at least one inference solution, the second inference solution set is obtained from the first inference solution set, the first inference solution set is a set generated based on the inference request, the first inference solution set includes at least one inference solution, and at least one of the confidence or risk value corresponding to each inference solution, each inference solution is used to achieve the reasoning task, the confidence is used to indicate the probability of achieving the reasoning task when using the inference solution, and the risk value is used to indicate the probability of changing the user experience when using the inference solution to achieve the reasoning task.

[0023] In the implementation manner of the present application, the producer generates the confidence and / or risk value of each reasoning scheme while generating the reasoning scheme. Therefore, when feeding back the reasoning scheme, the confidence and / or risk value of each reasoning scheme can be fed back at the same time, or the reasoning scheme can be screened based on the confidence of each reasoning scheme and then fed back, so that consumers can more accurately select the required scheme based on the received reasoning scheme.

[0024] In addition, the effects achieved by the second aspect and any optional implementation of the second aspect can refer to the effects achieved by the aforementioned first aspect or any optional implementation of the first aspect, and the similarities will not be repeated below.

[0025] In one possible embodiment, the aforementioned second set of inference schemes further includes: at least one of the confidence level or risk value corresponding to each inference scheme in at least one inference scheme. Therefore, in the embodiment of the present application, the producer can simultaneously provide feedback on the confidence level and / or risk value corresponding to each inference scheme, so that the consumer can receive data accurately based on the confidence level and / or risk value of each inference scheme and select the desired inference scheme.

[0026] In a possible implementation, the aforementioned reasoning request also carries a confidence indication and / or a risk indication, where the confidence indication is used to request provision of a confidence corresponding to the reasoning solution, and the risk indication is used to request provision of a risk value corresponding to the reasoning solution.

[0027] In a possible implementation, the confidence of the reasoning solutions included in the aforementioned second reasoning solution set is higher than a confidence threshold and / or the risk value is higher than a risk threshold.

[0028] In a possible implementation, the aforementioned inference request also carries a confidence threshold and / or a risk threshold.

[0029] In a possible implementation, the aforementioned second reasoning solution set includes one reasoning solution, and the one reasoning solution is screened from the first reasoning solution set according to the confidence level and / or the risk value.

[0030] In a possible implementation, the aforementioned inference request also carries a policy indication, where the policy indication is used to indicate that an inference solution is provided.

[0031] In a possible implementation, the aforementioned policy indication also indicates at least one of the filtering conditions, and the filtering conditions include at least one of the following: an inference scheme with the highest confidence, an inference scheme with the lowest risk value, or an inference scheme with the highest fusion value after fusing the confidence and risk values.

[0032] In a possible implementation, the aforementioned method may further include:

[0033] Sending a reasoning solution in a second reasoning solution set;

[0034] receiving an execution result, where the execution result is an execution result obtained by using a reasoning scheme in the second reasoning scheme set to accomplish the reasoning task;

[0035] The execution result is sent, and the execution result is used to update knowledge. The knowledge is used to determine a second set of reasoning solutions. The knowledge includes rules for completing the reasoning task.

[0036] In a third aspect, the present application provides a reasoning method, comprising:

[0037] The first device sends an inference request to the second device, where the inference request is used to request a solution for achieving a reasoning task, where the reasoning task includes a task of completing a user's intended goal;

[0038] The second device obtains a first inference solution set according to the inference request, where the first inference solution set includes at least one inference solution and at least one of a confidence level or a risk value corresponding to each inference solution. Each inference solution is used to achieve the inference task, the confidence level is used to indicate the probability of achieving the inference task when the inference solution is used, and the risk value is used to indicate the probability of changing the user experience when the inference solution is used to achieve the inference task.

[0039] The second device sends a second inference solution set to the first device, where the second inference solution set includes at least one solution in the first inference solution set.

[0040] In addition, the effects achieved by the third aspect and any optional implementation of the third aspect can refer to the effects achieved by the aforementioned first aspect or any optional implementation of the first aspect, and the similarities will not be repeated below.

[0041] In a possible implementation, the second device sends the second inference solution set to the first device, including:

[0042] The second device uses the first inference solution set as the second inference solution set;

[0043] The second device sends a second set of inference solutions to the first device.

[0044] In a possible implementation, the reasoning request also carries a confidence indication and / or a risk indication, where the confidence indication is used to request provision of a confidence level corresponding to the reasoning solution, and the risk indication is used to request provision of a risk value corresponding to the reasoning solution.

[0045] In a possible implementation, the second device sending the second inference solution set to the first device may include:

[0046] The second device filters out solutions with confidence levels higher than a confidence threshold and / or solutions with risk values ​​higher than a risk threshold from the first inference solution set to obtain a second inference solution set;

[0047] The second device sends a second set of inference solutions to the first device.

[0048] In a possible implementation, the aforementioned reasoning request also carries a confidence threshold and / or a risk threshold.

[0049] In a possible implementation, the second device sending the second inference solution set to the first device may further include:

[0050] The second device selects one of the inference schemes from the first inference scheme set as a scheme of the second inference scheme set according to at least one of a confidence level or a risk value corresponding to each inference scheme;

[0051] The second device sends a second set of inference solutions to the first device.

[0052] In one possible implementation, the second device selects one of the inference schemes from the first inference scheme set as a scheme of the second inference scheme set based on at least one of a confidence level or a risk value corresponding to each inference scheme, which may include:

[0053] The second device selects one of the inference schemes that meets the screening conditions from the first inference scheme set, where the screening conditions include at least one of the following: an inference scheme with the highest confidence, an inference scheme with the lowest risk value, or an inference scheme with the highest fusion value after fusing the confidence and risk values.

[0054] In a possible implementation, the aforementioned inference request further carries a policy indication, where the policy indication is used to indicate at least one of the screening conditions.

[0055] In a possible implementation, the aforementioned second reasoning solution set further includes at least one of a confidence level or a risk value corresponding to the reasoning solution.

[0056] In a possible implementation, the second device obtains the first reasoning solution set according to the reasoning request, including:

[0057] The second device obtains monitoring data and a knowledge base corresponding to the reasoning task, where the monitoring data includes data of devices related to the reasoning task, and the knowledge in the knowledge base includes rules for completing the reasoning task;

[0058] The second device obtains a first reasoning solution set based on the monitoring data and knowledge.

[0059] In a possible implementation, the aforementioned method may further include:

[0060] The first device sends an execution result to the second device, where the execution result is obtained after executing the inference scheme included in the second inference scheme set;

[0061] The second device updates the knowledge according to the execution result to obtain updated knowledge.

[0062] In a possible implementation, the aforementioned reasoning task includes: at least one of a fault location task, a fault repair task, a configuration verification task, a hidden danger identification task, or a solution recommendation task. The fault location task indicates locating the root cause of the equipment failure, the fault repair task indicates repairing the equipment that caused the failure, the configuration verification task indicates verifying the configuration of the equipment, the hidden danger identification task indicates obtaining the probability of the equipment causing a failure, and the solution recommendation task indicates obtaining at least one solution for executing the target task.

[0063] In a fourth aspect, the present application provides an inference device that can be used to execute the steps executed by the inference device in the first aspect and any possible implementation manner provided in the first aspect.

[0064] In one possible design, the present application may divide the inference device into functional modules according to the first aspect and any possible implementation method provided in the first aspect. For example, each functional module may be divided according to each function, or two or more functions may be integrated into a single processing module.

[0065] For example, the present application may divide the inference device into a transceiver module and a processing module according to function. The description of the possible technical solutions and beneficial effects performed by each of the above-mentioned functional modules can refer to the technical solutions provided in the above-mentioned first aspect or its corresponding possible implementation methods, and will not be repeated here.

[0066] In another possible design, the reasoning device includes: a memory and a processor, the memory and the processor being coupled. The memory is used to store computer instructions, and the processor is used to call the computer instructions to execute the method provided in the first aspect or its corresponding possible implementation manner.

[0067] In a fifth aspect, the present application provides an inference device that can be used to execute the steps executed by the inference device in the second aspect and any possible implementation manner provided in the second aspect.

[0068] In one possible design, the present application may divide the inference device into functional modules according to the second aspect and any possible implementation method provided in the second aspect. For example, each functional module may be divided according to each function, or two or more functions may be integrated into a single processing module.

[0069] For example, the present application may divide the inference device into a transceiver module and a processing module according to function. The description of the possible technical solutions and beneficial effects performed by each of the above-mentioned functional modules can refer to the technical solutions provided in the above-mentioned second aspect or its corresponding possible implementation methods, and will not be repeated here.

[0070] In another possible design, the reasoning device includes: a memory and a processor, the memory and the processor being coupled. The memory is used to store computer instructions, and the processor is used to call the computer instructions to execute the method provided in the second aspect or its corresponding possible implementation.

[0071] In the sixth aspect, the present application provides a communication system, which may also be referred to as a communication network, etc. The communication system may include multiple inference devices, at least one of which may be used to execute the method steps in the aforementioned second aspect or any optional embodiment of the second aspect, and at least one remaining inference device may also be used to execute the method steps in the aforementioned first aspect or any optional embodiment of the first aspect.

[0072] In a seventh aspect, the present application provides a computer-readable storage medium, such as a non-transitory computer-readable storage medium, having a computer program (or instruction) stored thereon, which, when executed on a computer device, causes the computer device to perform the method provided in the first aspect or its corresponding possible implementation manner, or the method provided in the second aspect or its corresponding possible implementation manner.

[0073] In an eighth aspect, the present application provides a computer program product which, when run on a computer device, enables the method provided in the first aspect or its corresponding possible implementation manner to be executed, or enables the method provided in the second aspect or its corresponding possible implementation manner to be executed.

[0074] In the ninth aspect, the present application provides a chip system, comprising: a processor, the processor being used to call and run a computer program stored in a memory from the memory, execute the method provided in the first aspect or its corresponding possible implementation manner, or execute the method provided in the second aspect or its corresponding possible implementation manner.

[0075] It can be understood that any of the systems, devices, computer storage media, computer program products or chip systems provided above can be applied to the corresponding methods provided in the first aspect or the second aspect.

[0076] For the specific implementation steps of the fourth to ninth aspects of this application and various possible implementation methods, as well as the beneficial effects brought about by each possible implementation method, please refer to the description of the various possible implementation methods in the first or second aspect, and will not be repeated here one by one. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] FIG1 is a schematic diagram of a system architecture provided by this application;

[0078] FIG2 is a schematic diagram of another system architecture provided by this application;

[0079] FIG3 is a schematic diagram of another system architecture provided by this application;

[0080] FIG4 is a schematic diagram of another system architecture provided by this application;

[0081] FIG5 is a flow chart of an inference method provided by this application;

[0082] FIG6 is a flow chart of another reasoning method provided by this application;

[0083] FIG7 is a schematic diagram of the structure of a data and knowledge base provided by this application;

[0084] FIG8 is a flow chart of another reasoning method provided by this application;

[0085] FIG9 is a flow chart of another reasoning method provided by this application;

[0086] FIG10 is a flow chart of another reasoning method provided by this application;

[0087] FIG11 is a flow chart of another reasoning method provided by this application;

[0088] FIG12 is a flow chart of another reasoning method provided by this application;

[0089] FIG13 is a flow chart of another reasoning method provided by this application;

[0090] FIG14 is a schematic diagram of the structure of an inference device provided by the present application;

[0091] FIG15 is a schematic diagram of the structure of another inference device provided by the present application;

[0092] FIG16 is a schematic diagram of the structure of another inference device provided in this application. DETAILED DESCRIPTION

[0093] The following will describe the technical solutions in the embodiments of this application in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0094] Referring to Figure 1, the network operation and maintenance architecture based on 5G communication technology can generally include three layers, namely: network element (NE), element management system (EMS), and network management system (NMS).

[0095] Among them, NE refers to various network devices that make up the network, which may include but are not limited to base stations, access points, various core network devices, etc. Optionally, NE can be a physical entity device or a virtual network function, which is not limited in this application. EMS can not only manage NE, but also be a provider of network operation and maintenance management services, and can perform specific network operation and maintenance management service operations according to the service call request sent by NMS. As the invoker of network operation and maintenance management services, NMS can send service call requests to EMS according to the customer's network operation and maintenance management needs, so as to call the network operation and maintenance management services in EMS that meet the customer's operation and maintenance management needs. Both EMS and NMS are composed of at least one device. The devices in EMS can be called network element management devices, and the devices in NMS can be called network management devices.

[0096] Network operation and maintenance includes network optimization, which typically includes improving user experience, reducing network energy consumption, or increasing network capacity. This is not specifically limited in the embodiments of the present application, and network optimization requirements and goals can be set based on actual needs. Traditional optimization solutions based on expert experience struggle to achieve satisfactory results in increasingly complex communication scenarios. To provide customers with a better network experience and improve operation and maintenance efficiency, the introduction of artificial intelligence for network operation and maintenance has become a mainstream trend. Machine learning is a branch of artificial intelligence. Machine learning uses algorithms to analyze data, learn from it, and then make decisions and predictions about real-world events. In network optimization scenarios, the decisions and predictions made by machine learning models are optimization recommendations, i.e., changes to the values ​​of parameters of devices in the network (sometimes referred to as parameter values ​​or simply parameters in the embodiments of the present application). Taking wireless base stations as an example, these parameters include scheduling parameters, resource management parameters, radio frequency (RF) parameters, power parameters, and handover parameters.

[0097] For example, in some scenarios, cognitive reasoning service consumers create fault location cognitive reasoning service tasks, including information such as the fault location, fault-related alarms, and fault events. After receiving the task, producers infer the fault chain and infer the root cause of the fault. The producer then returns the inference results to the consumer. Due to the complexity of network system topology and the dependency of fault propagation, there is a high probability of multiple root cause inference results. However, faced with multiple results, users cannot determine which root cause is the true one, which affects the overall root cause location and fault repair time, and thus affects the user experience.

[0098] For example, in some scenarios, cognitive reasoning service consumers create reasoning service requests for energy-saving solution recommendations, including the energy-saving application area, energy-saving goals and experience constraints, and the application time period. After receiving the request, producers perform reasoning and return a solution. However, cognitive reasoning service consumers often receive multiple sets of reasoning solutions from producers, and users may not be able to select the most appropriate solution based on their needs. Random selection leads to low problem-solving efficiency.

[0099] Therefore, the present application provides a reasoning method that generates a confidence level and / or a risk value for each reasoning solution while generating the reasoning solution, thereby enabling the user to select an appropriate solution based on their needs.

[0100] The system architecture and method provided by this application are introduced in more detail below.

[0101] The reasoning method provided in this application can be applied to a variety of communication networks, such as various optical networks or wireless networks. For example, refer to Figure 2, which is a schematic diagram of the architecture of another communication system provided in an embodiment of the present application. The communication network includes an NMS, an EMS, and network elements. In Figure 2, NMS is used to indicate the equipment included in the NMS, and EMS is used to indicate the equipment included in the EMS. This will not be repeated below. The relevant descriptions of the NMS, EMS, and network elements can be understood by referring to the relevant descriptions in Figure 1. The NMS is responsible for the management of the operation, management, and maintenance functions of the network. The EMS, or network element management system, is responsible for managing one or more categories of network elements. A network element is a network unit. In a wireless access network, it refers to a base station or base station controller, including but not limited to: one or more base stations in the global system for mobile communications (GSM), the universal mobile telecommunications system (UMTS), the long term evolution technology (LTE), and the new radio (NR), as well as base station controllers of GSM and UMTS. In the bearer network, it refers to one or more of the following: access router (ACC), metro edge gateway (MEG), metro aggregation edge gateway (MAEG), metro backbone router (MBB), 5G backbone router (5G BB), etc. In the core network, it refers to one or more of the following: mobile switching center (MSC), serving general packet radio system support node (SGSN), equipment identity register (EIR), packet data network gateway for user plane (PGW-U), access and mobility management function (AMF), unified data management (UDM), etc.

[0102] Based on the communication network described above, the solution provided in the embodiment of the present application can be jointly executed by the NMS and EMS, or jointly executed by the EMS and network elements, and can also be deployed in other devices.

[0103] For example, as shown in Figure 3, another architecture diagram provided by this application is shown. For EMS, NMS, and network elements, please refer to the relevant descriptions of Figure 1 above, which will not be repeated here.

[0104] The NMS may send an inference request to the EMS, where the inference request may be used to request the EMS to perform an inference solution for a maintenance or optimization task of the communication network.

[0105] After receiving the inference request, the EMS sends an alternative inference solution to the NMS. The EMS can collect data from network elements in advance, and the network elements can upload their own data or the data of the devices they manage to the EMS. The EMS can then generate an inference solution based on the collected data.

[0106] After receiving the alternative inference solutions, the NMS determines the final inference solution and sends it to the network element through the EMS.

[0107] After receiving the final inference plan, the network element can execute the final inference plan and feed back the execution result to the NMS.

[0108] Please refer to FIG4 , which is another schematic diagram of the architecture provided by this application.

[0109] Furthermore, NMS can be divided into intent service producers (such as the user's reasoning device or user interface, etc.) and task reasoning service consumers (such as NMS physical server or controller, etc.), and EMS can include cognitive reasoning service producers (such as EMS server or server cluster) and knowledge base (such as database or file system, etc.).

[0110] Cognitive reasoning service consumer: can be used to call the cognitive reasoning service provided by the cognitive reasoning service producer and send service requests to the cognitive reasoning service producer.

[0111] Cognitive Reasoning Service Producer: Provides cognitive reasoning services and returns reasoning solutions to cognitive reasoning service consumers. Cognitive reasoning service producers can be deployed within the EMS and also maintain a knowledge base. Using this knowledge base, producers have the ability to build and update knowledge.

[0112] In an embodiment of the present application, the intent service producer can be used to generate intent service requirements, such as maintenance intentions or optimization intentions for the communication network, and send the intentions to the cognitive reasoning service consumer, which sends the intent service requirements to the cognitive reasoning service producer.

[0113] The cognitive reasoning service producer maintains the knowledge base. Specifically, it can use the data collected from the network elements to update the knowledge in the knowledge base and use the knowledge in the knowledge base to complete the reasoning task, that is, generate reasoning solutions for the intent service needs and feed back the generated reasoning solutions as alternative reasoning solutions to the cognitive reasoning service consumers.

[0114] After the cognitive service reasoning consumer sends the final reasoning plan to the network element, it receives the execution result of the network element and feeds back the intent execution result to the intent service producer.

[0115] When performing network operations and maintenance (O&M), the solutions used can be data-driven, knowledge-driven, or a combination of both. As shown in Table 1, data-driven solutions have the advantage of not requiring precise modeling and enabling continuous learning and evolution. However, their interpretability is limited, requiring high-quality, large-scale data and the powerful computing power of the equipment. Knowledge-driven solutions have the advantages of strong interpretability and high execution efficiency, but they are typically costly to acquire, making continuous learning and evolution difficult. Furthermore, their knowledge mechanisms are unclear for large-scale, complex problems.

[0116] Therefore, based on the above analysis, this application can combine knowledge-driven and data-driven, and use their respective advantages to form a new technology route for the fusion of knowledge and data (i.e., cognitive reasoning). Combined with the business scenarios of autonomous driving networks, knowledge is mainly used as a carrier of expert experience to express some expert experience in operation and maintenance (such as fault propagation knowledge and configuration constraint knowledge). Data mainly reflects the current network status. Typical applications of cognitive reasoning of knowledge + data fusion in ADN include fault location (such as hardware failures in communication systems), fault repair (such as switching, restarting, rollback, and parameter adjustment), configuration verification, hidden danger identification (such as which devices may cause failures), solution recommendations (such as energy-saving solution recommendations), etc.

[0117] Table 1

[0118] The machine learning models mentioned in this application may include artificial intelligence models such as neural networks. The performance of most neural networks relies on a large amount of training data. Taking the network optimization scenario as an example to illustrate this, real-world network scenarios are complex and dynamically changing, and many factors affect network performance. In addition to the parameters of network equipment, the surrounding geographical environment (topography, surrounding buildings, etc.), user distribution and behavior, and even climate can affect network performance. This results in a huge state space and decision space for machine learning models, requiring a large amount of data as training data for learning. Alternatively, it can be understood that due to the numerous factors that affect network performance, machine learning models need to learn a lot of knowledge during the iterative training process. That is, machine learning models need to obtain a large amount of relationship between parameter changes and performance changes (or network state changes) as training data to learn the relationship between network configuration parameters and network performance changes. It should be noted that the models in this application include neural network models, and this application does not limit the specific type of model. For example, the models in this application can be convolutional neural network models, recurrent neural network models, deep residual network models, etc., and this will not be repeated below.

[0119] The following describes the method flow provided by this application in conjunction with the aforementioned system architecture.

[0120] Referring to FIG5 , a flowchart of an inference method provided by the present application is shown as follows.

[0121] 501. The NMS sends an inference request to the EMS.

[0122] Among them, the reasoning request is used to request the provision of a reasoning solution for achieving a reasoning task (or it can also be called a reasoning intention or cognitive reasoning service request, etc.). The reasoning task may include a task to complete the user's intended goal. The reasoning task may specifically include a task or intention to manage the communication network, such as a network optimization intention or a network maintenance intention.

[0123] The inference request may specifically come from the NMS or other management device that manages the communication network. For example, when the management device obtains the user's intention to manage the communication network, it may generate an inference request and send it to the EMS.

[0124] Optionally, the reasoning task may include but is not limited to one or more of the following: at least one of a fault location task, a fault repair task, a configuration verification task, a hidden danger identification task or a solution recommendation task. The fault location task indicates locating the root cause of the device's fault, the fault repair task indicates repairing the device that caused the fault, the configuration verification task indicates verifying the configuration of the device, the hidden danger identification task indicates obtaining the probability of the device causing a fault, and the solution recommendation task indicates obtaining at least one solution for executing the target task.

[0125] 502. The EMS obtains a first reasoning solution set based on the reasoning request.

[0126] After receiving the inference request, the EMS may generate one or more inference solutions, which are referred to as a first inference solution set for easy distinction.

[0127] Among them, when generating an inference plan, EMS can synchronously generate at least one of a confidence level or a risk value. The confidence level can be used to indicate the probability of achieving the inference task when using the inference plan, and the risk value is used to indicate the probability of changing the user experience when using the inference plan to achieve the inference task. The user experience can be specifically reflected through indicators such as the user service level agreement (SLA) and the user service quality (QoS).

[0128] Optionally, when generating one or more reasoning schemes, the reasoning schemes can be generated by combining data-driven and / or knowledge-driven methods. For example, monitoring data corresponding to the reasoning task and knowledge stored in a knowledge base can be obtained. The monitoring data can include data of devices related to the reasoning task, and the knowledge can include rules for completing the reasoning task. Subsequently, a first set of reasoning schemes is obtained based on the monitoring data and the knowledge. Therefore, in the embodiments of the present application, the reasoning scheme can be generated by combining the data actually generated by the devices in the communication network and the pre-acquired knowledge, thereby obtaining a more accurate reasoning scheme.

[0129] 503. The EMS sends the second inference solution set to the NMS.

[0130] After generating the first reasoning solution set, the EMS may feed back one or more reasoning solutions to the NMS as alternative reasoning solutions for the reasoning task, that is, the second reasoning solution set may include one or more reasoning solutions in the first reasoning solution set.

[0131] In the embodiments of the present application, when the EMS generates an inference solution, it can also generate a corresponding confidence level and / or risk value for each inference solution. Therefore, when feeding back alternative inference solutions to the NMS, it can also simultaneously feed back the confidence level and / or risk value of each inference solution, or select a solution that meets the user's needs, such as a solution with a higher confidence level and / or a lower risk level, allowing the user to more accurately select the desired inference solution.

[0132] Therefore, when the EMS feeds back the second inference solution set to the NMS, there are multiple situations, including but not limited to the following:

[0133] 1. Feedback all inference solutions and the confidence and / or risk value of each solution

[0134] After the EMS obtains the first set of inference solutions, it can directly use this set as the second set of inference solutions and feed this set of inference solutions back to the NMS. In other words, the inference solutions in the first set of inference solutions, along with the confidence and / or risk value of each solution, can be directly fed back to the NMS, allowing the user to select the desired inference solution based on the confidence and / or risk value of each solution.

[0135] Accordingly, the inference request may also include a confidence indicator and / or a risk indicator. The confidence indicator can be used to request the provision of a confidence level corresponding to the inference solution, and the risk indicator can be used to request the provision of a risk value corresponding to the inference solution. Of course, when the EMS provides feedback on the registered inference solution, it may also default to providing feedback to the NMS on the confidence level and / or risk value corresponding to each inference solution.

[0136] Therefore, when selecting the final inference solution, users can more accurately choose the desired solution based on the confidence and / or risk value of each inference solution, thereby improving the selected solution's compliance with user needs. Taking energy saving as an example, in conservative scenarios, users tend to choose solutions with low risk (even if the confidence level is not the highest), while in aggressive scenarios, users tend to choose solutions with high confidence (even if the risk is also high). For example, deep sleep of the remote radio unit (RRU) has good energy-saving effects and high confidence, but also high risk.

[0137] 2. Feedback one or more reasoning solutions with a confidence level higher than the confidence threshold and / or a risk value lower than the risk threshold

[0138] After the EMS obtains the first set of inference solutions, it can filter out one or more inference solutions with a confidence level higher than a confidence threshold and / or a risk level lower than a risk threshold and feed them back to the NMS. This allows the NMS to directly feed back inference solutions that meet its needs, allowing it to directly select the solution that meets its needs.

[0139] Optionally, the confidence level and / or risk value of each of the one or more inference solutions can be synchronously fed back. This allows the NMS to obtain the specific confidence level and / or risk value of each inference solution, thereby knowing in advance the expected results when executing the inference solution. Furthermore, when the EMS provides feedback on multiple inference solutions, it can more accurately select the desired inference solution based on the confidence level and / or risk value of each inference solution.

[0140] Optionally, the NMS may also carry a confidence threshold and / or a risk value threshold in the inference request to instruct the EMS to provide an inference solution with a confidence higher than the confidence threshold and / or a risk value lower than the risk value threshold.

[0141] Of course, in some scenarios, the cognitive reasoning service producer can also set a default confidence threshold and / or risk value threshold, so that the confidence threshold and / or risk value threshold does not need to be carried in the reasoning request to reduce the amount of data transmission and thus reduce communication overhead.

[0142] In this embodiment, the EMS can directly provide feedback to the NMS on inference solutions with confidence levels above a confidence threshold and / or risk levels below a risk threshold. This is equivalent to performing a preliminary screening and providing inference solutions that meet user needs. Therefore, the inference solutions provided to users are all those that meet their needs, providing users with more usable inference solutions.

[0143] 3. Feedback a reasoning solution that meets the user's specified requirements

[0144] In the implementation manner of the present application, in order to further reduce the selection decisions that the user needs to make, an inference solution that meets the user's specified requirements can be directly fed back based on the user's needs, so that the user does not need to make a second selection and communication overhead can be reduced.

[0145] After the EMS obtains the first set of inference solutions, it can filter out a single inference solution that meets the selection criteria and feed it back to the NMS. Specifically, one inference solution can be selected based on the confidence and / or risk value of each inference solution and fed back to the NMS as part of the second set of inference solutions. Therefore, users can select a received inference solution as the final inference solution without having to select it.

[0146] The screening condition may specifically include but is not limited to at least one of the following: an inference scheme with the highest confidence, an inference scheme with the lowest risk value, or an inference scheme with the highest fusion value after fusing the confidence and risk values.

[0147] For example, EMS can feedback a reasoning solution with higher confidence, or feedback a reasoning solution with higher confidence, or feedback a solution that meets the requirements by combining confidence and risk value, such as the highest-ranked solution with both confidence and risk values ​​ranked in the top N, or the reasoning solution with the highest value obtained after fusion calculation of confidence and risk value, etc.

[0148] Optionally, the NMS may carry a policy indication in the inference request, and the policy indication may be used to indicate a screening condition, so that the EMS may screen out one of the one or more generated inference solutions based on the policy indication and feed the result back to the NMS.

[0149] Of course, the EMS can select one of the generated one or more inference solutions based on pre-set screening conditions and feed it back to the NMS. The specific conditions can be determined according to the actual application scenario, and this application does not limit this.

[0150] Therefore, in the implementation manner of the present application, an inference solution can be directly screened out and fed back to the NMS, so that the user can directly obtain an inference solution that meets the needs, and the user can directly determine the final inference solution without re-screening.

[0151] The above is a general introduction to the method flow provided by this application. The following is a more detailed introduction to the method flow provided by this application in combination with the above method flow and application scenarios.

[0152] In addition, after receiving one or more candidate inference solutions, the NMS can continue to send the final inference solution to the network element. After obtaining the execution results of the final inference solution, the data and knowledge base are updated so that the actual feedback of the execution results of the inference solution can be combined to improve the accuracy of the next inference.

[0153] Referring to FIG6 , a flowchart of another reasoning method provided by the present application is described below.

[0154] 601. EMS collects data from network elements and updates data and knowledge base.

[0155] Among them, EMS can collect information related to various scenarios from network elements, such as data generated or collected by various devices. After the data is collected, the data and knowledge base can be updated.

[0156] In the embodiments of the present application, when the EMS performs reasoning to generate a reasoning solution, a combination of data-driven and knowledge-driven methods can be used to generate the reasoning solution. Accordingly, during the data collection phase, the collected data can be used to update the data and knowledge base, and the data and knowledge base can be used to perform subsequent reasoning steps.

[0157] Specifically, when collecting data, you can collect data generated by the operation of each device itself, information about each device connected to it, and information about the environment in which each device is located. For example, you can collect device fault alarm information, device energy consumption information, device configuration information, or device access information.

[0158] For example, as shown in FIG7 , a knowledge base can be established that includes a variety of pre-set knowledge. This knowledge can typically include expert experience or knowledge learned from collected data. Knowledge can be used to indicate the rules for various types of functions or tasks, or the logical links between the types of data that each function relies on. In embodiments of the present application, the knowledge base can include knowledge of the rules for completing reasoning tasks.

[0159] When generating inference plans, data-driven approaches can also be incorporated. Therefore, machine learning can be pre-trained based on collected data to generate a machine learning model, such as the neural network shown in Figure 7. When generating inference plans, the collected device data can be used as input to the neural network, which outputs the inference plan. Different neural networks can be set for different inference tasks, or a large cross-modal model can be set up to implement multiple functions.

[0160] It should be noted that step 601 can be executed when each device is initialized, or it can be actively reported when the data of the network element changes, or the EMS can collect data periodically. The specific timing can be determined according to the actual application scenario. This application does not limit the execution timing of step 601.

[0161] 602. The NMS sends an inference request to the EMS.

[0162] The reasoning request can be used to request EMS to provide a reasoning solution for achieving a reasoning task (or also called a reasoning intention or cognitive reasoning service request, etc.). The reasoning task may include a task or intention to manage the communication network, such as a task or intention to maintain or optimize the communication network, etc.

[0163] The inference request can carry information about the inference task, allowing the EMS to identify the inference task service to be executed. For example, it can carry the task name, task fields, or task identifier of the inference task. It can also carry task description information. For example, for a fault location task, the task description information that can be carried may include the fault occurrence time period, fault location, and fault event content.

[0164] In addition, the NMS can actively send an inference request to the EMS, or send an inference request to the EMS when triggered by other devices or users. The specific limitation can be based on the actual application scenario, and this application does not limit it.

[0165] For the three different scenarios mentioned above, the inference request may carry different content. For details, please refer to the corresponding introductions of Figures 8 to 10 below, which will not be repeated here.

[0166] 603. The EMS obtains one or more reasoning solutions and the confidence and / or risk value of each solution according to the reasoning request.

[0167] After receiving the reasoning request, EMS can obtain one or more alternative reasoning schemes and the confidence and / or risk value of each scheme based on the reasoning request in combination with data-driven and / or knowledge-driven methods, that is, the aforementioned first reasoning scheme set.

[0168] Specifically, when generating an inference scheme, one or more inference schemes can be generated based on data-driven, such as using the data of the equipment in the communication network that needs to be maintained or optimized as the input of the neural network, thereby outputting the inference scheme and the confidence and / or risk value corresponding to each inference scheme. Alternatively, one or more inference schemes can be generated based on knowledge-driven, such as reading the data of the equipment in the communication network that needs to be maintained or optimized, determining the maintenance or optimization scheme corresponding to each equipment based on knowledge, thereby forming a maintenance or optimization scheme for the communication network and the confidence and / or risk value corresponding to each inference scheme. Inference schemes can also be generated by combining data-driven and knowledge-driven; such as using the collected data of the equipment as the input of the neural network, using knowledge as a constraint, and outputting one or more inference schemes and the confidence and / or risk value corresponding to each inference scheme, etc. The specific calculation method can be adjusted according to the actual application scenario, and this application does not limit this.

[0169] Typically, the confidence value is related to the probability of achieving the reasoning task. This application exemplarily introduces the positive correlation between the confidence value and the probability of achieving the reasoning task. That is, the higher the confidence, the higher the probability of achieving the reasoning task when the reasoning scheme is used, and the lower the confidence, the lower the probability of achieving the reasoning task when the reasoning scheme is used. Of course, in some scenarios, it can also be replaced with a negative correlation between the confidence value and the probability of achieving the reasoning task, and this application is not limited to this.

[0170] Accordingly, the risk value is related to the probability of changing the user service level agreement (SLA) when the reasoning scheme is used to perform the reasoning task. This application exemplarily introduces the example of the positive correlation between the risk value and the probability of changing the agreement SLA when the reasoning scheme is used to perform the reasoning task. That is, the lower the risk value, the lower the probability of changing the SLA when the reasoning scheme is used to perform the reasoning task, and the higher the risk value, the higher the probability of changing the SLA when the reasoning scheme is used to perform the reasoning task. Of course, it can also be replaced with a negative correlation between the risk value and the probability of changing the agreement SLA when the reasoning scheme is used to perform the reasoning task. The specific replacement can be based on the actual application scenario, and this application does not limit this.

[0171] 604. The EMS sends one or more alternative inference solutions to the NMS.

[0172] After generating one or more reasoning schemes and the confidence and / or risk value of each scheme, EMS can send one or more alternative reasoning schemes, namely the aforementioned second reasoning scheme set, to NMS based on the one or more reasoning schemes and the confidence and / or risk value of each scheme.

[0173] Optionally, when sending alternative reasoning schemes, the confidence and / or risk value of each alternative reasoning scheme may be sent. For example, when sending multiple alternative reasoning schemes, the confidence and / or risk value of each reasoning scheme may be sent synchronously; or the generated reasoning schemes may be screened based on the confidence and / or risk value of each reasoning scheme, and one or more schemes that meet the user's needs may be screened for feedback. In this case, the confidence and / or risk value of each reasoning scheme may be sent synchronously, or the confidence and / or risk value of each reasoning scheme may not be sent synchronously; or, if only one reasoning scheme is fed back, the confidence and / or risk value of each reasoning scheme may be sent synchronously, or the confidence and / or risk value of each reasoning scheme may not be sent synchronously.

[0174] Correspondingly, for the different feedback methods mentioned above, the inference request may carry different content. For details, please refer to the corresponding introductions of Figures 8 to 10 below, which will not be repeated here.

[0175] 605. The NMS sends the final inference solution to the network element.

[0176] If the NMS receives an alternative reasoning scheme sent by the EMS, it can directly send the alternative reasoning scheme as the final reasoning scheme to the network element; if the NMS receives multiple alternative reasoning schemes sent by the EMS, it can filter out one of the multiple alternative reasoning schemes as the final reasoning scheme.

[0177] Optionally, if the NMS receives multiple alternative inference schemes sent by the EMS, the NMS can randomly select an inference scheme; if the confidence and / or risk value corresponding to each inference scheme is received at the same time, one of the inference schemes can be selected from the multiple inference schemes based on the confidence and / or risk value of each inference scheme, such as selecting the inference scheme with the highest confidence or the scheme with the lowest risk value.

[0178] In the embodiment of the present application, the EMS can provide the NMS with feedback on alternative reasoning solutions based on the confidence level and / or risk value of the reasoning solution, so that the NMS can more accurately determine a more optimal reasoning solution.

[0179] In addition, when NMS sends the final inference plan to the network element, the EMS can usually forward the inference plan and send the maintenance or optimization plan for each device in the communication network included in the inference plan to each network element device, such as instructing each network element to execute a specified configuration, thereby improving the performance of the communication network. The sending method of this final inference plan will not be repeated in the following implementation methods of this application.

[0180] 606. The network element feeds back the execution result to the NMS.

[0181] After each network element executes the inference plan, the execution results can be fed back to the NMS, such as the operation status of the network element after fault repair, or the energy consumption of the network element after adopting the energy-saving plan.

[0182] 607. The NMS sends the execution result to the EMS.

[0183] After receiving the execution result, the NMS may send the execution result to the EMS.

[0184] 608. The EMS updates the data or knowledge base based on the execution results.

[0185] After receiving the execution result, EMS can update the data or knowledge base based on the execution result, so that EMS can obtain a more accurate reasoning solution when generating a reasoning solution for executing reasoning next time.

[0186] In the implementation manner of the present application, when generating an inference scheme, the EMS can synchronously generate the confidence and / or risk value of each inference scheme, that is, it can feedback alternative inference schemes based on the confidence and / or risk value of each inference scheme, such as screening the generated inference schemes based on the confidence and / or risk value of each inference scheme and then issuing them, or synchronously downloading the confidence and / or risk value of each inference scheme, etc., so that the NMS can obtain the required inference scheme more accurately.

[0187] For ease of understanding, the following introduces different feedback strategies for inference schemes.

[0188] Solution 1: Feedback on all inference solutions and the confidence and / or risk value of each solution

[0189] Referring to FIG8 , a flowchart of another reasoning method provided by the present application is shown as follows.

[0190] 801. EMS collects data from network elements and updates data and knowledge base.

[0191] For step 801, please refer to the related description of the aforementioned step 601, which will not be repeated here.

[0192] 802. The NMS sends an inference request to the EMS. The inference request may carry a confidence indication and / or a risk indication.

[0193] When the NMS needs the EMS to provide an inference solution for executing an inference task, the NMS can send an inference request to the EMS. The inference request can refer to the relevant description of the aforementioned step 602, and similarities are not repeated here.

[0194] Among them, optionally, the inference request can carry a confidence indication (confidence reporting indication) and / or a risk indication (risk reporting indication). The confidence indication can be used to instruct the EMS to synchronously feedback the confidence of each inference scheme when providing an inference scheme, and the risk indication can be used to instruct the EMS to synchronously feedback the risk value of each inference scheme when providing an inference scheme.

[0195] 803. The EMS obtains one or more reasoning solutions and the confidence and / or risk value of each solution.

[0196] For step 803, please refer to the related description of the aforementioned step 603, which will not be repeated here.

[0197] 804. The EMS sends one or more alternative reasoning solutions and the confidence and / or risk value of each solution to the NMS.

[0198] After the EMS generates one or more reasoning solutions and the confidence and / or risk value of each solution, the one or more reasoning solutions and the confidence and / or risk value of each solution may be sent to the NMS.

[0199] 805. The NMS sends the final inference solution to the network element.

[0200] If the NMS receives only one alternative reasoning scheme, it can determine whether to use the alternative reasoning scheme as the final reasoning scheme based on the confidence and / or risk value of the alternative reasoning scheme. For example, if the confidence of the alternative reasoning scheme exceeds a first threshold or the risk value is lower than a second threshold, the alternative reasoning scheme can be used as the final reasoning scheme. Of course, the alternative reasoning scheme can also be directly used as the final reasoning scheme, which can be adjusted according to the actual application scenario. In addition, the final reasoning scheme can also be called the actual reasoning scheme, the target reasoning scheme, or other interchangeable names, etc., which are not limited by this application.

[0201] If the NMS receives multiple alternative reasoning schemes, it can select one of the reasoning schemes as the final reasoning scheme based on the confidence and / or risk value of the multiple alternative reasoning schemes. For example, an inference scheme with the highest confidence level may be selected as the final reasoning scheme, or an inference scheme with the lowest risk value may be selected as the final reasoning scheme, or an inference scheme with the lowest risk value may be selected as the final reasoning scheme among inference schemes with confidence levels higher than a third threshold, or an inference scheme with the highest confidence level may be selected among inference schemes with risk values ​​lower than a fourth threshold, etc. The specific selection can be made based on actual user needs, and this application does not limit this.

[0202] For example, when the EMS sends an inference plan to the NMS, there are multiple situations. The method of selecting the final inference plan may also be different for different situations:

[0203] (1) For only confidence returned: NMS can choose the solution with the highest confidence;

[0204] (2) For only returning risk: NMS can choose the solution with the lowest risk;

[0205] (3) Regarding the returned confidence and risk: NMS can customize the selection strategy (such as confidence priority, risk value priority, or weighting confidence and risk and then selecting based on the weighted value, etc.).

[0206] 806. The network element feeds back the execution result to the NMS.

[0207] 807. The NMS sends the execution result to the EMS.

[0208] 808. The EMS updates the data or knowledge base based on the execution results.

[0209] Among them, steps 806 to 808 can refer to the introduction of the aforementioned steps 606 to 608, and will not be repeated here.

[0210] Therefore, in the embodiment of the present application, when EMS provides an inference scheme to NMS, it can also simultaneously provide the confidence and / or risk value of each inference scheme, so that when NMS selects the final inference scheme, it can more accurately select an inference scheme that meets the needs based on the confidence and / or risk value of each inference scheme.

[0211] Implementation Option 2: Feedback is consistent with one or more reasoning schemes with a confidence level higher than a confidence threshold and / or a risk value lower than a risk value threshold. Referring to FIG. 9 , a flowchart of another reasoning method provided by the present application is described below.

[0212] 901. EMS collects data from network elements and updates data and knowledge base.

[0213] For step 901, please refer to the relevant description of the aforementioned step 601, which will not be repeated here.

[0214] 902. The NMS sends an inference request to the EMS. The inference request may carry a confidence threshold and / or a risk threshold.

[0215] The inference request can refer to the related description of the aforementioned step 602, and similarities are not repeated here.

[0216] The difference is that the reasoning request can optionally carry a confidence threshold and / or a risk threshold. The confidence threshold can be used to instruct the EMS to provide reasoning solutions with a confidence level higher than the confidence threshold, and the risk threshold can be used to instruct the EMS to provide reasoning solutions with a risk value lower than the risk threshold.

[0217] 903. The EMS obtains one or more reasoning solutions and the confidence and / or risk value of each solution.

[0218] Among them, step 903 can refer to the introduction of the aforementioned step 603, which will not be repeated here.

[0219] 904. The EMS sends one or more candidate reasoning solutions to the NMS, and the confidence of each solution exceeds a confidence threshold and / or the risk value is lower than a risk threshold.

[0220] After the EMS generates one or more inference schemes, it can screen out inference schemes with confidence exceeding a confidence threshold and / or risk value lower than a risk threshold from the one or more inference schemes as alternative inference schemes, and send the one or more alternative inference schemes to the NMS.

[0221] The confidence threshold and / or risk threshold may come from an inference request, or may be a pre-set default value, or a value received when the inference plan was last generated, etc. It may be adjusted according to the actual application scenario, and this application does not limit this.

[0222] Optionally, when EMS sends one or more alternative reasoning schemes to NMS, it may also simultaneously send the confidence and / or risk value of each scheme, so that NMS can know the possible effects of using each alternative reasoning scheme to achieve the reasoning task.

[0223] 905. The NMS sends the final inference solution to the network element.

[0224] After receiving the one or more candidate reasoning solutions sent by the EMS, the NMS may select one of the one or more candidate reasoning solutions as the final reasoning solution.

[0225] Specifically, when the NMS receives an alternative reasoning solution, it can directly use the alternative reasoning solution as the final reasoning solution; when the NMS receives multiple alternative reasoning solutions, it can randomly select an alternative reasoning solution as the final reasoning solution.

[0226] Optionally, if the EMS simultaneously sends the confidence and / or risk value of each alternative reasoning scheme when sending it to the NMS, then when the NMS selects a final reasoning scheme from multiple alternative reasoning schemes, it can select the final reasoning scheme based on the confidence and / or risk value of each scheme. For example, a reasoning scheme with the highest confidence level is selected as the final reasoning scheme, or a reasoning scheme with the lowest risk value is selected as the final reasoning scheme, or a reasoning scheme with the lowest risk value is selected as the final reasoning scheme among the reasoning schemes with confidence levels higher than the third threshold, or a reasoning scheme with the highest confidence level is selected among the reasoning schemes with risk values ​​lower than the fourth threshold, etc. The specific selection can be made based on actual user needs, and this application does not limit this.

[0227] 906. The network element feeds back the execution result to the NMS.

[0228] 907. The NMS sends the execution result to the EMS.

[0229] 908. The EMS updates the data or knowledge base based on the execution results.

[0230] Among them, steps 906 to 908 can refer to the introduction of the aforementioned steps 606 to 608, and will not be repeated here.

[0231] Therefore, in the embodiments of the present application, when the EMS provides an inference solution to the NMS, it can provide an inference solution with a confidence level exceeding the confidence threshold and / or a risk value below the risk threshold, thereby providing an inference solution that meets the user's needs, allowing the user to more accurately filter out the inference solution that meets their needs. This is equivalent to expressing the user's preferences for confidence and risk through confidenceThld (confidence threshold) and riskThld (risk threshold), and can filter out inference results that do not meet the user's preferences in advance, reducing the number of returned solutions, reducing the burden on users to check the results, and improving efficiency.

[0232] Implementation 3: Feedback a reasoning solution that meets the user's specified requirements

[0233] Referring to FIG10 , a flowchart of another reasoning method provided by the present application is shown as follows.

[0234] 1001. EMS collects data from network elements and updates data and knowledge base.

[0235] For step 1001, please refer to the relevant description of the aforementioned step 601, which will not be repeated here.

[0236] 1002. The NMS sends an inference request to the EMS. The inference request may carry a policy indication.

[0237] The inference request can refer to the related description of the aforementioned step 602, and similarities are not repeated here.

[0238] The difference is that the inference request can optionally carry a policy indication to instruct the EMS to provide an inference solution selected according to the policy indicated by the policy indication. For example, the policy indication can be confidence priority, risk value priority, or a fusion of confidence and risk value, etc. The specific determination can be based on the actual application scenario and is not limited by this application.

[0239] 1003. The EMS obtains one or more reasoning solutions and the confidence and / or risk value of each solution.

[0240] Among them, step 1003 can refer to the introduction of the aforementioned step 603, which will not be repeated here.

[0241] 1004. The EMS sends an alternative reasoning solution to the NMS, where the solution is a solution selected based on the policy indicated by the policy indication.

[0242] After the EMS obtains one or more inference solutions, it can select one of them as an alternative inference solution based on the policy instruction. The policy instruction can come from the inference request, a pre-set policy instruction, or an instruction received during the last inference solution screening. The specific method can be determined based on the actual application scenario and is not limited by this application.

[0243] Specifically, the strategy indicated by the strategy indication may include, but is not limited to: confidence priority, risk value priority, or a fusion of confidence and risk value indications. Confidence priority means that the EMS can select an alternative reasoning scheme based on the confidence of each reasoning scheme, such as selecting the reasoning scheme with the highest confidence as the alternative reasoning scheme; risk value priority means that the EMS can select an alternative reasoning scheme based on the risk value of each reasoning scheme, such as selecting the scheme with the lowest risk value as the alternative reasoning scheme; and fusion of confidence and risk value indications means that the EMS can select an alternative reasoning scheme based on the confidence and risk value of each reasoning scheme, such as weighted fusion of the confidence and risk value of each reasoning scheme, and selecting an alternative reasoning scheme based on the fusion value. Optionally, the risk value is generally negatively correlated with its probability of being selected. For example, the lower the risk value, the higher the probability of being selected, and the higher the risk value, the lower the probability of being selected, thereby selecting an alternative reasoning scheme with a lower risk value.

[0244] 1005. The NMS sends the final inference solution to the network element.

[0245] In this embodiment, the NMS can receive an alternative inference solution from the EMS. Therefore, the NMS does not need to make a selection and can directly send the alternative inference solution to the EMS as the final inference solution. This is equivalent to the EMS having already screened the alternative inference solutions based on the policy instructions when sending them. This reduces the NMS's screening steps and improves the NMS's efficiency in selecting the final inference solution.

[0246] 1006. The network element feeds back the execution result to the NMS.

[0247] 1007. The NMS sends the execution result to the EMS.

[0248] 1008. The EMS updates the data or knowledge base based on the execution results.

[0249] Among them, steps 1006 to 1008 can refer to the introduction of the aforementioned steps 606 to 608, and will not be repeated here.

[0250] In the implementation manner of the present application, the EMS can directly send the reasoning solution that meets the requirements to the NMS based on the policy indication. The NMS can directly obtain the reasoning solution that meets the requirements without further screening, thereby obtaining the required reasoning solution more accurately.

[0251] The following is an illustrative introduction to possible application scenarios of the method provided in this application in combination with some possible application scenarios.

[0252] Application Scenario 1: Fault Location and Repair

[0253] The fault location cognitive reasoning service task is to locate hardware faults or software faults in the communication network, which may include but is not limited to the fault location, fault-related alarms, fault events and other information. The fault repair cognitive reasoning service task may include but is not limited to fault handling methods, such as restarting, adjusting parameters, master-slave switching or resetting the board.

[0254] Referring to FIG11 , a flowchart of another reasoning method provided by the present application is described below.

[0255] 1101. Cognitive reasoning service producers collect data.

[0256] Among them, the cognitive reasoning service producer can collect data in advance, which may specifically include information about the network elements it manages, such as the logs, configurations, status or detected events of each network element.

[0257] For example, the EMS can collect information such as alarms, events, traffic statistics, statistical information, packet capture data, configuration information, or operation logs generated by network elements managed by the EMS. Alarms can include alarms generated when a device detects a fault; traffic statistics can include statistical traffic information; packet capture data includes data intercepted by the device during operation, which is usually used to test network security; configuration information can include information about the configuration used by each device during operation; and operation logs include records of each device's operations during active or passive operations.

[0258] After receiving multiple requests, the cognitive reasoning service producer collects data generated by each device, or actively reports the changed data to the cognitive reasoning service producer after the device data changes, or after the execution result is generated after the previous execution of the reasoning task, the knowledge is stored in the form of triples, such as adding confidence attributes or risk attributes to the knowledge. If the use of certain knowledge ultimately leads to problems when the plan is issued, the risk of the corresponding knowledge will be increased.

[0259] 1102. The cognitive reasoning service consumer creates a fault location and repair service request to the cognitive reasoning service producer under the triggering of the intent service producer.

[0260] Typically, cognitive reasoning services may be generated by an intent service producer and trigger cognitive reasoning service consumers to create fault location and repair service requests to the cognitive reasoning service producer.

[0261] The request may include but is not limited to:

[0262] Task name: such as fault location, fault repair

[0263] Task description information: fault object (FailurePredictionObjec), potential fault type (PotentialFailureType), fault occurrence time (EventTime), fault severity, fault location, fault event content, etc.

[0264] The specific information carried in the request may also be referred to the records in the relevant standards. For example, the standard document 3GPP TS28.104 V1.1.0 records the information carried in the request in the fault location scenario.

[0265] In addition, in combination with the aforementioned embodiments 1 to 3, the indications that may be carried in the inference request may include, but are not limited to, one or more of the following:

[0266] 1. Reporting a confidence indication (i.e., the aforementioned confidence indication) and / or reporting a risk indication (i.e., the aforementioned risk indication);

[0267] 2. confidenceThld (i.e., the aforementioned confidence threshold) and / or riskThld (i.e., the aforementioned risk threshold);

[0268] 3. sortPolicy (i.e. the aforementioned policy indication).

[0269] Among them, reporting the confidence indication can be used to instruct the cognitive reasoning service producer to provide fault location and repair solutions while also providing feedback on the confidence level (confidence) of each solution. Reporting the risk indication can be used to instruct the cognitive reasoning service producer to provide fault location and repair solutions while also providing feedback on the risk value (risk) of each reasoning solution.

[0270] confidenceThld can be used to instruct cognitive reasoning service producers to feedback solutions with a confidence level not lower than or higher than confidenceThld when providing fault location and repair solutions; riskThld can be used to instruct cognitive reasoning service producers to feedback solutions with a risk value lower than or no higher than riskThld when providing fault location and repair solutions.

[0271] sortPolicy can be used to instruct cognitive reasoning service producers to select a solution for feedback when providing fault location and repair solutions according to the strategy indicated by sortPolicy. For example, sortPolicy can be set to confidencePriority (confidence priority), riskPriority (low risk priority), or a custom weighted calculation formula of confidence and risk, such as: sortPolicy = 0.5*confidence + 0.5*(1-risk).

[0272] In the implementation mode of this application, by enhancing the interface between the cognitive reasoning service producer and the cognitive reasoning service consumer, the confidence and risk information of the cognitive reasoning request return field is supplemented, assisting the users on the cognitive reasoning service consumer side to better select and judge solutions, thereby improving the efficiency of the problem closure loop.

[0273] 1103. The cognitive reasoning service producer performs fault location and repair service reasoning, generates and selects reasoning solutions that meet the requirements.

[0274] After receiving a request to create a fault location and repair service, the cognitive reasoning service producer can retrieve relevant knowledge and data based on the request, including the confidence and risk information in the knowledge triples. It then executes a cognitive reasoning algorithm or runs a neural network, calculating the confidence and / or risk of the fault root cause and repair action during the reasoning process, and generates a fault location and repair service reasoning solution. This fault location and repair reasoning solution may include:

[0275] Specifically, when executing fault location and repair service reasoning, it can be completed through the producer's internal algorithms. For example, when reasoning about the root cause, the probability of the root cause can be calculated through algorithms such as random walks. This approach requires enhancing the algorithmic capabilities within the cognitive reasoning producer; alternatively, the knowledge in the knowledge base can be combined to generate a fault location and repair service reasoning solution.

[0276] For example, in a knowledge base, knowledge can be stored in triples. The knowledge base itself carries confidence and risk information. By enriching the triple attribute definitions in the knowledge graph, confidence and risk attributes can be added to the traditional head entity-relationship-tail entity triple. As shown in Figure 12, for a certain fault phenomenon, the probability of causing alarm 1 is 0.6, the probability of causing alarm 2 is 0.2, and the probability of causing KPI x anomaly is 0.8. If KPI x anomaly occurs, the probability of causing alarm 3 is 0.7, the probability of causing alarm 4 is 0.6, and so on. In the case of alarm 1, the risk of repairing the problem through restart is 0.5, and so on.

[0277] 1104. The cognitive reasoning service producer feeds back one or more alternative fault location and repair solutions to the cognitive reasoning service consumer.

[0278] When returning a fault location and repair service reasoning solution, the specific returned solution may be determined based on the instructions carried in the reasoning request.

[0279] Such situations can be divided into but not limited to the following:

[0280] 1. Regarding the confidence indicator and / or risk indicator carried in the inference request: When feeding back the inference solution, the confidence and / or risk are synchronously returned, for example, it can be expressed as: [{"solution":x,"confidence":0.5,"risk":0.7},...];

[0281] 2. For confidenceThld and / or riskThld carried in the inference request: when feeding back the inference solution, the generated solutions can be filtered to select those with confidence > (≥) confidenceThld and / or risk < (≤) riskThld;

[0282] 3. For inference requests containing sortPolicy: Based on different policies, you can filter out an inference solution from the generated solutions based on the sortPolicy. For example, if sortPolicy = confidencePriority, you can filter out the inference solution with the highest confidence for feedback; if sortPolicy = riskPriority, you can directly return the solution with the lowest risk; if sortPolicy is a custom weighted fusion formula, you can calculate according to that formula and return an inference solution based on the calculated score, such as the solution with the highest score.

[0283] In the embodiments of the present application, corresponding inference schemes can be adaptively returned according to different indications carried in the inference request. Therefore, in multiple situations, the requester can accurately select the required scheme, so as to achieve more accurate fault location and repair.

[0284] For the aforementioned situation 1, when the cognitive inference service consumer sends a cognitive inference service request, in addition to the existing basic task information description, a confidence reporting indication and a risk reporting indication are added. The cognitive inference service producer returns the inference result and the corresponding confidence and risk according to the request content, indicating the probability that the result is correct and the probability that the result may cause risks respectively. This enables users to have a clearer understanding of the reliability of each scheme, select the scheme they need, and improve the efficiency of problem closure. That is, when the cognitive inference service producer returns multiple inference results, the confidence and / or risk information of the results is carried, indicating the confidence and risk of the results, helping users better select a suitable scheme according to their own needs. That is, when the cognitive inference service producer returns multiple inference results, the confidence and / or risk information of the results is carried, indicating the confidence and risk of the results, helping users better select a suitable scheme according to their own needs.

[0285] For the aforementioned situation 2, when the cognitive inference service consumer sends a request, the request content includes fields of confidenceThld (confidence threshold) and / or riskThld (risk threshold). The cognitive inference service producer can perform pruning in advance based on the threshold requirements during the inference process, accelerating the inference efficiency. At the same time, only the results that meet confidence > confidenceThld AND risk < riskThld are returned, reducing the number of invalid results returned and reducing the burden on the consumer to select a scheme. That is, by using confidenceThld and riskThld to express the user's preferences for confidence and risk, the inference results that do not meet the user's preferences can be filtered out in advance, reducing the number of returned schemes, reducing the burden on the user to check the results, and improving the efficiency.

[0286] For situation 3, a scheme selection strategy sortPolicy is added. When sending a cognitive inference service request, the user's preferred selection strategy is carried. The cognitive inference service producer directly completes the selection of multiple schemes internally, and the cognitive inference service consumer directly receives the optimal scheme, saving the time for the consumer to select and analyze separately and improving the efficiency of problem closure. That is, the user-defined scheme selection strategy field sortPolicy is carried in the cognitive inference service request, and the cognitive inference service producer directly returns only one optimal result according to the user's preference, eliminating the need to return multiple results for the consumer to select.

[0287] 1105. The cognitive reasoning service consumer sends the final fault location and repair plan to the network element.

[0288] After receiving one or more fault location and repair solutions sent by the cognitive reasoning service producer, the cognitive reasoning service consumer selects a fault location and repair solution as the final fault location and repair solution, and sends the final fault location and repair solution to the relevant network elements.

[0289] Generally, if a cognitive reasoning service consumer receives a fault location and repair solution, the fault location and repair solution can be used as the final fault location and repair solution and sent to the corresponding network element.

[0290] If a cognitive reasoning service consumer receives multiple fault location and repair solutions, it can select one from the multiple solutions as the final solution. This selection can be done randomly or according to pre-set rules. For example, if the confidence and / or risk of each solution are received simultaneously, the solution with the highest or lowest confidence can be selected, or the confidence and risk can be weighted and fused to select the solution with the highest or lowest value based on the fused score. The specific selection can be adjusted according to the actual application scenario.

[0291] In some possible scenarios, when multiple inference solutions are received, the intent service producer may select one of the inference solutions as the final inference solution. For example, in the fault location and repair solution of the embodiment of the present application, the intent service producer may select the final fault location and repair solution.

[0292] Furthermore, cognitive reasoning service consumers can directly deliver received fault location and repair solutions to network elements, or they can generate different control data based on different network elements before delivering them. For example, based on the fault location results, fault repair data or configuration can be delivered to relevant network elements to implement repair after fault location.

[0293] 1106. The network element feeds back the repair result to the cognitive reasoning service consumer.

[0294] After receiving the fault location and repair plan from the cognitive reasoning service consumer, the network element can execute the corresponding plan to achieve fault repair. For example, fault repair can be achieved through switching, restarting, rollback, or parameter adjustment. Each network element can perform the corresponding operation and feedback the repair results to the cognitive reasoning service consumer, such as whether the fault has been repaired after executing the received repair plan.

[0295] 1107. The cognitive reasoning service consumer feeds back the repair result to the intent service producer.

[0296] After the cognitive reasoning service consumer receives the repair result from the network element, it can feedback the repair result to the intent service producer so that the intent service producer can know the repair status of the fault, or determine whether its intention has been achieved. If not, the aforementioned step 1102 can be re-executed.

[0297] 1108. The cognitive reasoning service consumer sends the repair result to the cognitive reasoning service producer.

[0298] The cognitive reasoning service consumer may also send a repair result to the cognitive reasoning service producer to instruct the cognitive reasoning service producer to update the knowledge base.

[0299] 1109. The cognitive reasoning service producer updates the knowledge base.

[0300] After receiving the execution results of the reasoning solution, the cognitive reasoning service producer can update the knowledge base. Based on the final execution results, the cognitive reasoning service consumer sends a request to the cognitive reasoning service producer to update the knowledge base. Generally, if the intended goal is ultimately achieved based on certain knowledge in the knowledge base, the confidence of this knowledge is high, and the confidence attribute of the knowledge triple can be increased accordingly. If the use of certain knowledge ultimately causes problems when the solution is delivered, the risk of the corresponding knowledge is increased.

[0301] Typically, knowledge-driven approaches require enhanced knowledge mining capabilities, supplementing the confidence and risk information in the knowledge base, and enabling reverse updates to the knowledge base based on the application results (consumer feedback). Referring to Figure 12, after a fault recovery plan is generated and executed in the network element, the knowledge in the knowledge base can be reversely modified based on the execution results, for example, by adjusting the confidence and / or risk associated with each fault recovery action for each alarm.

[0302] In the implementation manner of the present application, for the situation of fault repair and location, the cognitive reasoning service producer can generate a variety of feasible fault location and repair solutions, as well as the confidence and / or risk corresponding to each solution. When feeding back the fault and repair solutions to the cognitive reasoning service consumer, the confidence and / or risk corresponding to each solution can be fed back synchronously, so that the cognitive reasoning service consumer can select the required fault location and repair solution based on the confidence and / or risk corresponding to each solution. Furthermore, the cognitive reasoning service producer can filter the generated fault location and repair solutions based on the confidence and / or risk corresponding to each solution, such as by confidenceThld and / or riskThld, or by sortPolicy, so that solutions that meet their needs can be fed back to the cognitive reasoning service consumer, reducing the steps that the cognitive reasoning service consumer needs to select, or eliminating the need for the cognitive reasoning service consumer to make a selection, etc., so that the cognitive reasoning service can obtain the required solution more directly and accurately.

[0303] Application scenario 2: Recommended energy-saving scenarios

[0304] Referring to FIG13 , a flowchart of another reasoning method provided by the present application is described below.

[0305] 1301. Cognitive reasoning service producers collect data.

[0306] Among them, step 1301 is similar to the aforementioned step 1101. Please refer to the relevant description of the aforementioned step 1101 for details, which will not be repeated here.

[0307] For energy-saving recommendation scenarios, the data usually collected by cognitive reasoning service producers may include but are not limited to: network element operating performance, quality of experience (QoE) data, network element configuration data or network analysis data, etc. The required data can be obtained based on the actual application scenario.

[0308] 1302. The cognitive reasoning service consumer creates an energy-saving decision cognitive reasoning service request to the cognitive reasoning service producer under the triggering of the intention service producer.

[0309] Similar to the aforementioned step 1102, the similarities are not repeated here, and the differences are introduced below.

[0310] The create energy-saving decision cognitive reasoning service request may carry the energy-saving effect area, energy-saving target and experience constraint or effect time period, etc.

[0311] 1303. The cognitive reasoning service producer performs energy-saving decision-making service reasoning to generate energy-saving solutions.

[0312] After the producer of the reasoning service receives the request for the cognitive reasoning service to create an energy-saving decision, it can perform energy-saving decision service reasoning based on the information carried in the request, and generate an energy-saving plan based on knowledge-driven and / or data-driven.

[0313] The energy-saving plan may include a plan for energy-saving management of devices requiring energy-saving management within the energy-saving area or energy-saving time period mentioned in the request. Specifically, the energy-saving plan may include, but is not limited to, information such as the objects / network elements to be energy-saving, the energy-saving type, network load trends, energy-saving recommendations (energy-saving strategies, such as shutting down certain network elements during energy-saving periods or within energy-saving areas), or the grid status after energy-saving.

[0314] 1304. The cognitive reasoning service producer feeds back one or more energy savings to the cognitive reasoning service.

[0315] 1305. The cognitive reasoning service consumer sends the final energy-saving plan to the network element.

[0316] 1306. The network element feeds back the energy-saving results to the cognitive reasoning service consumer.

[0317] 1307. The cognitive reasoning service consumer feeds back the energy-saving results to the intention service producer.

[0318] 1308. The cognitive reasoning service consumer sends the energy saving result to the cognitive reasoning service producer.

[0319] 1309. Update the knowledge base.

[0320] Among them, steps 1304 to 1309 are similar to the aforementioned steps 1104 to 1109, except that the aforementioned fault location and repair are replaced by energy saving. For details, please refer to the relevant descriptions of the aforementioned steps 1104 to 1109, which will not be repeated here.

[0321] In the implementation manner of the present application, for the energy-saving recommendation scenario, the cognitive reasoning service producer can still generate one or more energy-saving solutions (or energy-saving recommendation solutions or energy-saving decision solutions, etc.), as well as the confidence and / or risk value corresponding to each energy-saving solution, and feedback the generated energy-saving solution and the confidence and / or risk value corresponding to each energy-saving solution based on the instructions of the cognitive reasoning service consumer, so that the cognitive reasoning service consumer can screen the solutions based on the confidence and / or risk value of each solution to accurately obtain the required energy-saving solution. Alternatively, the cognitive reasoning service producer can be instructed to screen the generated energy-saving solutions and then feedback, so that the cognitive reasoning service consumer can directly and accurately obtain the required energy-saving solution without screening.

[0322] To facilitate understanding of the effects achieved by the solution provided in this application, a comparative explanation is provided below combining existing solutions with the solution provided in this application.

[0323] The following is a comparison between the existing solutions and the solution provided in this application from the two dimensions of NMS and EMS.

[0324] See Table 2 for a comparison of the steps performed by the EMS in the existing scheme and the scheme provided in this application.

[0325] Table 2

[0326] See Table 3 for a comparison of the steps performed by the NMS in the existing solution and the solution provided in this application.

[0327] Table 3

[0328] Obviously, through the solutions provided by the embodiments of this application, cognitive reasoning service consumers can obtain more accurate reasoning solutions. For example, they can filter based on the confidence and / or risk of each received reasoning solution to obtain a reasoning solution that meets their needs; or they can carry instructions in the reasoning request, such as confidenceThld / riskThld / sortPolicy, so that cognitive reasoning service producers can filter the generated solutions, allowing cognitive reasoning service consumers to directly obtain a reasoning solution that meets their needs.

[0329] The above describes the method flow provided by this application. The following describes the device for executing the above method.

[0330] Referring to FIG14 , a schematic diagram of the structure of an inference device provided by the present application is described as follows.

[0331] The reasoning device may be deployed in the aforementioned EMS or cognitive reasoning service producer, and may include:

[0332] The transceiver module 1401 is configured to receive an inference request, wherein the inference request is configured to request a solution for achieving a reasoning task, wherein the reasoning task includes a task for completing a user's intended goal;

[0333] Processing module 1402 is configured to obtain a first inference solution set based on the inference request, the first inference solution set including at least one inference solution and at least one of a confidence level or a risk value corresponding to each inference solution, wherein each inference solution is used to achieve the inference task, the confidence level is used to indicate a probability of achieving the inference task when the inference solution is used, and the risk value is used to indicate a probability of a change in user experience when the inference solution is used to achieve the inference task;

[0334] The transceiver module 1401 is further configured to send a second reasoning solution set, where the second reasoning solution set includes at least one solution in the first reasoning solution set.

[0335] In a possible implementation, the processing module 1402 is further configured to use the first reasoning solution set as the second reasoning solution set;

[0336] The transceiver module 1401 is specifically configured to send the second inference solution set.

[0337] In a possible implementation, the inference request also carries a confidence indication and / or a risk indication, wherein the confidence indication is used to request provision of a confidence corresponding to the inference solution, and the risk indication is used to request provision of a risk value corresponding to the inference solution.

[0338] In a possible implementation, the processing module 1402 is specifically configured to filter out, from the first set of inference solutions, solutions having a confidence level higher than a confidence threshold and / or solutions having a risk value higher than a risk threshold, to obtain the second set of inference solutions;

[0339] The transceiver module 1401 is specifically configured to send the second inference solution set.

[0340] In a possible implementation, the inference request also carries the confidence threshold and / or the risk threshold.

[0341] In a possible implementation, the processing module 1402 is specifically configured to select one of the inference schemes from the first inference scheme set as a scheme of the second inference scheme set according to at least one of a confidence level or a risk value corresponding to each inference scheme;

[0342] The transceiver module 1401 is specifically configured to send the second inference solution set.

[0343] In one possible implementation, the processing module 1402 is specifically used to filter out one of the inference schemes that meets the filtering conditions from the first inference scheme set, and the filtering conditions include at least one of the following: an inference scheme with the highest confidence, an inference scheme with the lowest risk value, or an inference scheme with the highest fusion value after fusing the confidence and risk values.

[0344] In a possible implementation, the inference request further carries a policy indication, where the policy indication is used to indicate at least one of the screening conditions.

[0345] In a possible implementation, the second reasoning scheme set further includes at least one of a confidence level or a risk value corresponding to the reasoning scheme.

[0346] In one possible implementation, the processing module 1402 is specifically used to: obtain monitoring data and a knowledge base corresponding to the reasoning task, the monitoring data including data of equipment related to the reasoning task, and the knowledge in the knowledge base including rules for completing the reasoning task; and obtain the first reasoning scheme set based on the monitoring data and the knowledge.

[0347] In a possible implementation, the transceiver module 1401 is further configured to obtain an execution result, where the execution result is obtained after executing the reasoning scheme included in the second reasoning scheme set;

[0348] The processing module 1402 is further configured to update the knowledge according to the execution result to obtain updated knowledge.

[0349] In one possible implementation, the reasoning task includes: at least one of a fault location task, a fault repair task, a configuration verification task, a hidden danger identification task, or a solution recommendation task. The fault location task indicates locating the root cause of the device's fault, the fault repair task indicates repairing the device that caused the fault, the configuration verification task indicates verifying the configuration of the device, the hidden danger identification task indicates obtaining the probability of the device causing a fault, and the solution recommendation task indicates obtaining at least one solution for executing the target task.

[0350] Referring to FIG15 , a schematic diagram of the structure of an inference device provided by the present application is described as follows.

[0351] The reasoning device may be deployed in the aforementioned NMS or cognitive reasoning service consumer, and may include:

[0352] The transceiver module 1501 is configured to send an inference request, wherein the inference request is configured to request a solution for achieving a reasoning task, wherein the reasoning task includes a task for completing a user's intended goal;

[0353] The transceiver module 1501 is also used to receive a second reasoning scheme set, which includes at least one reasoning scheme. The second reasoning scheme set is obtained from the first reasoning scheme set. The first reasoning scheme set is a set generated based on the reasoning request. The first reasoning scheme set includes at least one reasoning scheme, and at least one of the confidence or risk value corresponding to each reasoning scheme. Each reasoning scheme is used to achieve the reasoning task, the confidence is used to indicate the probability of achieving the reasoning task when using the reasoning scheme, and the risk value is used to indicate the probability of changing the user experience when using the reasoning scheme to achieve the reasoning task.

[0354] In a possible implementation, the second reasoning scheme set further includes: at least one of a confidence level or a risk value corresponding to each reasoning scheme in the at least one reasoning scheme.

[0355] In a possible implementation, the inference request also carries a confidence indication and / or a risk indication, wherein the confidence indication is used to request provision of a confidence corresponding to the inference solution, and the risk indication is used to request provision of a risk value corresponding to the inference solution.

[0356] In a possible implementation, the confidence of the reasoning solutions included in the second reasoning solution set is higher than a confidence threshold and / or the risk value is higher than a risk threshold.

[0357] In a possible implementation, the inference request also carries the confidence threshold and / or the risk threshold.

[0358] In a possible implementation, the second reasoning solution set includes one reasoning solution, and the one reasoning solution is obtained by screening from the first reasoning solution set according to the confidence level and / or the risk value.

[0359] In a possible implementation, the inference request further carries a policy indication, where the policy indication is used to indicate providing an inference solution.

[0360] In a possible implementation, the policy indication further indicates at least one of the filtering conditions, and the filtering condition includes at least one of the following: an inference scheme with the highest confidence, an inference scheme with the lowest risk value, or an inference scheme with the highest fusion value after fusing the confidence and risk values.

[0361] In a possible implementation, the transceiver module 1501 is further configured to:

[0362] Sending a reasoning solution in the second reasoning solution set;

[0363] receiving an execution result, where the execution result is an execution result obtained by using a reasoning scheme in the second reasoning scheme set to achieve the reasoning task;

[0364] The execution result is sent, where the execution result is used to update knowledge, where the knowledge is used to determine the second set of reasoning solutions, and the knowledge includes rules for completing the reasoning task.

[0365] As shown in Figure 16, it is a schematic diagram of the hardware structure of a computer device 160 provided in an embodiment of the present application. The computer device 160 can be used to implement the functions of the NMS or EMS in the above method.

[0366] The computer device 160 shown in FIG16 may include a processor 1601 , a memory 1602 , a communication interface 1603 , and a bus 1604 . The processor 1601 , the memory 1602 , and the communication interface 1603 may be connected via the bus 1604 .

[0367] The processor 1601 is the control center of the computer device 160 and can be a general-purpose central processing unit (CPU) or other general-purpose processors, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0368] As an example, the processor 1601 may include one or more CPUs, such as CPU 0 and CPU 1 shown in FIG. 16 .

[0369] The memory 1602 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0370] In one possible implementation, memory 1602 may exist independently of processor 1601. Memory 1602 may be connected to processor 1601 via bus 1604 and used to store data, instructions, or program code. When processor 1601 calls and executes the instructions or program code stored in memory 1602, the method provided in the embodiments of the present application can be implemented.

[0371] In another possible implementation, the memory 1602 may also be integrated with the processor 1601 .

[0372] Communication interface 1603 is used to connect computer device 160 to other devices via a communication network, such as Ethernet, a radio access network (RAN), or a wireless local area network (WLAN). Communication interface 1603 may include a receiving unit for receiving data and a transmitting unit for sending data.

[0373] Bus 1604 can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. This bus can be classified as an address bus, a data bus, a control bus, etc. For ease of illustration, FIG16 shows only one thick line, but this does not mean that there is only one bus or only one type of bus.

[0374] It should be noted that the structure shown in FIG16 does not constitute a limitation on the computer device 160. In addition to the components shown in FIG16, the computer device 160 may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0375] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary general-purpose hardware, and of course can also be implemented by dedicated hardware including application-specific integrated circuits, dedicated CPUs, dedicated memories, dedicated components, etc. In general, all functions performed by computer programs can be easily implemented with corresponding hardware, and the specific hardware structures used to implement the same function can also be diverse, such as analog circuits, digital circuits, or dedicated circuits. However, for the present application, software program implementation is a better implementation method in most cases. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer's floppy disk, USB flash drive, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., including a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0376] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0377] A computer-readable storage medium is also provided in an embodiment of the present application, which stores a program for training a model or performing an inference task. When the program is run on a computer, the computer executes all or part of the steps in the method described in the embodiments shown in Figures 4 to 12 above.

[0378] The present application also provides a digital processing chip. The digital processing chip integrates circuitry and one or more interfaces for implementing the aforementioned processor or processor functions. When the digital processing chip integrates memory, it can perform the method steps of any one or more of the aforementioned embodiments. When the digital processing chip does not integrate memory, it can be connected to an external memory via a communication interface. The digital processing chip implements the method steps of any one or more of the aforementioned embodiments based on program code stored in the external memory.

[0379] A computer program product is also provided in the embodiment of the present application, and the computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0380] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium, which may include: ROM, RAM, disk or CD, etc.

[0381] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. The term "and / or" in this application is merely a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. The naming or numbering of the steps in this application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The named or numbered process steps can be changed in the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved. The division of modules in this application is a logical division. There may be other division methods when implementing in actual applications. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between each other shown or discussed can be through some ports, and the indirect coupling or communication connection between modules can be electrical or other similar forms, which are not limited in this application. In addition, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed in multiple circuit modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application.

Claims

1. A reasoning method, characterized in that: include: receiving a reasoning request, the reasoning request being used to request a solution for achieving a reasoning task, the reasoning task comprising a task for completing a user's intended goal; Acquire a first reasoning scheme set according to the reasoning request, the first reasoning scheme set including at least one reasoning scheme and at least one of a confidence level or a risk value corresponding to each reasoning scheme, each reasoning scheme is used to achieve the reasoning task, the confidence level is used to indicate a probability of achieving the reasoning task when the reasoning scheme is used, and the risk value is used to indicate a probability of changing a user experience when the reasoning scheme is used to achieve the reasoning task; A second set of inference solutions is sent, where the second set of inference solutions includes at least one solution in the first set of inference solutions.

2. The method according to claim 1, characterized in that The sending of the second reasoning solution set includes: Using the first reasoning solution set as the second reasoning solution set; The second set of inference solutions is sent.

3. The method according to claim 2, characterized in that The reasoning request also carries a confidence indication and / or a risk indication, wherein the confidence indication is used to request provision of a confidence corresponding to the reasoning solution, and the risk indication is used to request provision of a risk value corresponding to the reasoning solution.

4. The method according to claim 1, characterized in that: The sending of the second reasoning solution set includes: Filtering out the solutions whose confidences are higher than a confidence threshold and / or the solutions whose risk values ​​are higher than a risk threshold from the first reasoning solution set, to obtain the second reasoning solution set; The second set of inference solutions is sent.

5. The method according to claim 4, characterized in that The inference request also carries the confidence threshold and / or the risk threshold.

6. The method according to claim 1, characterized in that The sending of the second reasoning solution set further includes: From the first set of reasoning solutions, according to at least one of the confidence or risk value corresponding to each of the reasoning solutions, select one of the reasoning solutions as a solution of the second set of reasoning solutions; The second set of inference solutions is sent.

7. The method according to claim 6, characterized in that The selecting one of the inference schemes from the first inference scheme set according to at least one of the confidence level or the risk value corresponding to each inference scheme comprises: From the first set of inference schemes, one of the inference schemes that meets the screening conditions is screened out, and the screening conditions include at least one of the following: an inference scheme with the highest confidence, an inference scheme with the lowest risk value, or an inference scheme with the highest fusion value after integrating the confidence and the risk value.

8. The method according to claim 7, characterized in that The inference request also carries a policy indication, where the policy indication is used to indicate at least one of the screening conditions.

9. The method according to any one of claims 4 to 8, characterized in that: The second reasoning scheme set also includes at least one of a confidence level or a risk value corresponding to the reasoning scheme.

10. The method according to any one of claims 1 to 9, characterized in that The acquiring a first reasoning scheme set according to the reasoning request includes: Acquire monitoring data and a knowledge base corresponding to the reasoning task, wherein the monitoring data includes data of a device related to the reasoning task, and the knowledge in the knowledge base includes rules for completing the reasoning task; The first reasoning solution set is obtained according to the monitoring data and the knowledge.

11. The method according to claim 10, characterized in that The method further comprises: Obtaining an execution result, where the execution result is obtained after executing the reasoning scheme included in the second reasoning scheme set; The knowledge is updated according to the execution result to obtain updated knowledge.

12. The method according to any one of claims 1 to 11, characterized in that The reasoning task includes: at least one of a fault location task, a fault repair task, a configuration verification task, a hidden danger identification task or a solution recommendation task. The fault location task indicates locating the root cause of the device fault, the fault repair task indicates repairing the device that has caused the fault, the configuration verification task indicates verifying the configuration of the device, the hidden danger identification task indicates obtaining the probability of the device causing a fault, and the solution recommendation task indicates obtaining at least one solution for executing the target task.

13. A method of reasoning, characterized in that: include: Sending a reasoning request, wherein the reasoning request is used to request a solution for achieving a reasoning task, wherein the reasoning task includes a task for completing a user's intended goal; Receive a second reasoning scheme set, the second reasoning scheme set includes at least one reasoning scheme, the second reasoning scheme set is obtained from the first reasoning scheme set, the first reasoning scheme set is a set generated based on the reasoning request, the first reasoning scheme set includes at least one reasoning scheme, and at least one of a confidence level or a risk value corresponding to each reasoning scheme, each reasoning scheme is used to achieve the reasoning task, the confidence level is used to indicate the probability of achieving the reasoning task when using the reasoning scheme, and the risk value is used to indicate the probability of changing the user experience when using the reasoning scheme to achieve the reasoning task.

14. The method according to claim 13, characterized in that The second reasoning scheme set further includes: at least one of a confidence level or a risk value corresponding to each reasoning scheme in the at least one reasoning scheme.

15. The method according to claim 14, characterized in that The reasoning request also carries a confidence indication and / or a risk indication, wherein the confidence indication is used to request provision of a confidence corresponding to the reasoning solution, and the risk indication is used to request provision of a risk value corresponding to the reasoning solution.

16. The method according to claim 13, characterized in that The confidence of the reasoning solutions included in the second reasoning solution set is higher than a confidence threshold and / or the risk value is higher than a risk threshold.

17. The method according to claim 16, characterized in that The inference request also carries the confidence threshold and / or the risk threshold.

18. The method according to claim 13, characterized in that The second reasoning solution set includes one reasoning solution, and the one reasoning solution is screened from the first reasoning solution set according to the confidence level and / or the risk value.

19. The method according to claim 18, characterized in that The inference request also carries a policy indication, and the policy indication is used to indicate providing an inference solution.

20. According to the method of claim 19, the policy indication also indicates at least one of the filtering conditions, and the filtering condition includes at least one of the following: an inference scheme with the highest confidence, an inference scheme with the lowest risk value, or an inference scheme with the highest fusion value after integrating the confidence and the risk value.

21. The method according to any one of claims 13 to 20, characterized in that The method further comprises: Sending a reasoning solution in the second reasoning solution set; receiving an execution result, where the execution result is an execution result obtained by using a reasoning scheme in the second reasoning scheme set to accomplish the reasoning task; The execution result is sent, where the execution result is used to update knowledge, where the knowledge is used to determine the second reasoning solution set, and the knowledge includes rules for completing the reasoning task.

22. A method of reasoning, characterized in that: include: The first device sends a reasoning request to the second device, wherein the reasoning request is used to request a solution for achieving a reasoning task, wherein the reasoning task includes a task of completing a user's intended goal; The second device obtains a first reasoning scheme set according to the reasoning request, where the first reasoning scheme set includes at least one reasoning scheme and at least one of a confidence level or a risk value corresponding to each reasoning scheme, where each reasoning scheme is used to achieve the reasoning task, the confidence level is used to indicate a probability of achieving the reasoning task when the reasoning scheme is used, and the risk value is used to indicate a probability of changing a user experience when the reasoning scheme is used to achieve the reasoning task; The second device sends a second set of inference solutions to the first device, where the second set of inference solutions includes at least one solution in the first set of inference solutions.

23. The method according to claim 22, characterized in that The second device sends a second reasoning solution set to the first device, including: The second device uses the first reasoning solution set as the second reasoning solution set; The second device sends the second set of inference solutions to the first device.

24. The method according to claim 22, characterized in that The second device sends a second reasoning solution set to the first device, including: The second device selects, from the first set of inference solutions, solutions whose confidences are higher than a confidence threshold and / or solutions whose risk values ​​are higher than a risk threshold, to obtain the second set of inference solutions; The second device sends the second set of inference solutions to the first device.

25. The method according to claim 22, characterized in that The second device sends a second reasoning solution set to the first device, further comprising: The second device selects one of the inference schemes from the first inference scheme set as a scheme of the second inference scheme set according to at least one of the confidence or risk value corresponding to each inference scheme; The second device sends the second set of inference solutions to the first device.

26. An inference device, characterized in that: include: A transceiver module, used to perform the steps of receiving data or sending data as described in any one of claims 1 to 12; A processing module, configured to execute the step of processing data as claimed in any one of claims 1 to 12.

27. An inference device, characterized in that: include: A transceiver module, used to perform the steps of receiving data or sending data as described in any one of claims 13 to 21; A processing module, used to execute the step of processing data as described in any one of claims 13-21.

28. A communication system, characterized in that: The system includes a first device and a second device; The second device is used to perform the method according to any one of claims 13 to 21; The first device is configured to execute the method according to any one of claims 1 to 12.

29. An inference device, characterized in that: include: A memory storing executable program instructions; and A processor, wherein the processor is used to couple with the memory, read and execute instructions in the memory, and trigger the reasoning device to implement the method described in any one of claims 1 to 12 or 13 to 21.

30. A computer-readable storage medium comprising instructions, which, when executed on a computer, enables the computer to execute the method according to any one of claims 1 to 12, or to execute the method according to any one of claims 13 to 21.

Images