Abnormality processing method, device and system and medium
By integrating the exception reporting function in the operating system, using the exception prediction classification model and processing library, the rapid classification of exception events and the demonstration of self-service solutions is achieved, the problem of low exception handling efficiency in the existing technology is solved, and the operating system is improved Exception handling efficiency and convenience.
Patent Information
- Application Number
- CN202411193755.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-12-06
AI Technical Summary
In the prior art, the exception reporting and processing method of the operating system is separated from the system, resulting in low efficiency and learning time costs, making it difficult to handle exceptions accurately and efficiently.
By integrating the exception reporting function in the operating system, responding to the exception reporting instructions, receiving the exception event description information input from the agent, inputting it into the trained exception prediction classification model for classification, and matching the self-service solution in the exception processing library according to the classification results. It is also displayed on the terminal interface. If the matching fails, it will be pushed to the operation and maintenance terminal for processing priority sorting and manual intervention.
The integration of exception reporting and processing is achieved, which reduces learning time costs and improves the efficiency of exception processing. Through semantic analysis and self-service solutions, the operation convenience and processing efficiency of the agent are improved.
Smart Images

Figure CN119089306A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of financial technology, and in particular to an exception handling method, device, system and medium. Background Art
[0002] In financial companies such as property insurance companies, customer service representatives (agents) usually complete their daily work tasks through operating systems, such as handling insurance consultation, quotation, insurance, claims, etc. If there are abnormalities in the operating system, it will affect the daily work of the agents, so the abnormalities of the operating system need to be handled in a timely manner.
[0003] There are generally two ways to solve abnormal phenomena in operating systems. The first is to report the abnormal interface screenshot and abnormal description through the event reporting system, and then wait for the backend support staff to handle it; the second is to report the abnormality through internal communication channels such as the enterprise communication group, and handle it in real time in the communication group.
[0004] Although both of the above-mentioned two processing methods can handle exceptions, the exception reporting and processing paths of these two methods are separated from the operating system, which makes the efficiency of exception reporting low and has a learning time cost, making it difficult to accurately and efficiently handle the exceptions of the operating system. Summary of the invention
[0005] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide an exception handling method, device, system and medium that can be applied to financial technology or other related fields. Its main purpose is to improve the efficiency of exception reporting and processing of the operating system, facilitate seat operation, and reduce learning time costs.
[0006] The technical solution of the present invention is as follows:
[0007] A first aspect of the present invention provides an exception handling method, which is applied to an operating system, and the method comprises:
[0008] In response to the abnormality reporting instruction, receiving description information of the abnormal event input by the agent;
[0009] Inputting the description information of the abnormal event into the trained abnormal prediction classification model to perform abnormal classification on the abnormal event;
[0010] Matching is performed in an exception handling library according to the exception classification result, wherein the exception handling library stores resolved exceptions and corresponding solutions;
[0011] When the target anomaly is matched with the corresponding self-service solution, the self-service solution is displayed on the agent's terminal interface.
[0012] In one embodiment, after responding to the abnormality reporting instruction, the method further includes:
[0013] Displaying an exception reproduction prompt, wherein the exception reproduction prompt is used to prompt the agent to input an operation to reproduce the exception;
[0014] After detecting that the agent starts to input the operation to reproduce the exception, all DOM events when the exception occurs are recorded.
[0015] In one embodiment, the method further comprises:
[0016] When the target exception cannot be matched or the self-service solution is invalid, the exception event is marked as an unresolved exception;
[0017] Obtain all DOM events of the unresolved exception and push them to the operation and maintenance end, so as to replay the unresolved exception at the operation and maintenance end;
[0018] Receive and display a manual solution sent by the operation and maintenance end, where the manual solution is input by the operation and maintenance personnel based on the playback of the unresolved exception.
[0019] In one embodiment, the method further comprises:
[0020] Collect statistics on abnormal information within a specified time period, including the abnormal type, abnormal scope, and abnormal resolution time of the abnormal event;
[0021] The processing priorities of different exception types are determined based on the statistical results of the exception information.
[0022] In one embodiment, when the target exception cannot be matched or the self-service solution is invalid, after marking the exception event as an unresolved exception, the method further includes:
[0023] Determine the corresponding processing priority based on the exception classification results of each unresolved exception;
[0024] Generate a dynamic exception resolution plan based on the processing priority of all unresolved exceptions and push it to the operation and maintenance end.
[0025] In one embodiment, the method further comprises:
[0026] Receive exception resolution results from agents.
[0027] In one embodiment, after receiving the abnormality resolution result fed back by the agent, the method further includes:
[0028] confirming the matching degree between the abnormal classification result and the abnormal event according to the abnormal resolution result;
[0029] Feedback correction is performed on the abnormal prediction classification model according to the matching degree.
[0030] A second aspect of the present invention provides an exception handling device, comprising:
[0031] The abnormality reporting module is used to receive the description information of the abnormal event input by the seat in response to the abnormality reporting instruction;
[0032] A prediction and classification module, used to input the description information of the abnormal event into the trained abnormal prediction and classification model to perform abnormal classification on the abnormal event;
[0033] An exception matching module, used to match in an exception handling library according to the exception classification result, wherein the exception handling library stores resolved exceptions and corresponding solutions;
[0034] The solution display module is used to display the self-service solution on the terminal interface of the agent when the target exception is matched with the corresponding self-service solution.
[0035] A third aspect of the present invention provides an exception handling system, the system comprising at least one processor; and
[0036] a memory communicatively connected to the at least one processor; wherein,
[0037] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above exception handling method.
[0038] A fourth aspect of the present invention provides a non-volatile computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by one or more processors, the one or more processors can execute the above-mentioned exception handling method.
[0039] Beneficial effects: The present invention discloses an exception handling method, device, system and medium. Compared with the prior art, the embodiments of the present invention are applied to the operating system, including: in response to an exception reporting instruction, receiving the description information of the abnormal event input by the agent; inputting the description information of the abnormal event into the trained abnormal prediction classification model to classify the abnormal event; matching in the exception handling library according to the abnormal classification result, the exception handling library stores the resolved exceptions and corresponding solutions; when the target exception is matched with the corresponding self-service solution, the self-service solution is displayed on the terminal interface of the agent. By integrating the exception reporting with the operating system, the agent can report the description information of the abnormal event in the operating system with one click, which is convenient for the agent to operate and effectively saves the learning time cost. After the semantic analysis of the reported abnormal description information, the abnormal event is automatically classified and the matching self-service solution is fed back, thereby improving the efficiency of the operating system's exception handling. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:
[0041] Figure 1 A flow chart of an exception handling method provided by an embodiment of the present invention;
[0042] Figure 2 Another flow chart of the exception handling method provided by an embodiment of the present invention;
[0043] Figure 3 Another flow chart of the exception handling method provided by an embodiment of the present invention;
[0044] Figure 4 Another flow chart of the exception handling method provided by an embodiment of the present invention;
[0045] Figure 5 Another flow chart of the exception handling method provided by an embodiment of the present invention;
[0046] Figure 6 Another flow chart of the exception handling method provided by an embodiment of the present invention;
[0047] Figure 7 A schematic diagram of the functional modules of the exception handling device provided by an embodiment of the present invention;
[0048] Figure 8 A schematic diagram of the hardware structure of an exception handling system provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0049] In order to make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail below. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. The embodiments of the present invention are described below in conjunction with the accompanying drawings.
[0050] In financial companies such as property insurance companies, customer service representatives (agents) usually complete their daily work tasks through operating systems, such as handling insurance consultation, quotation, insurance, claims, etc. If there are abnormalities in the operating system, it will affect the daily work of the agents, so the abnormalities of the operating system need to be handled in a timely manner.
[0051] There are generally two ways to solve abnormal phenomena in operating systems. The first is to report the abnormal interface screenshot and abnormal description through the event reporting system, and then wait for the backend support staff to handle it; the second is to report the abnormality through internal communication channels such as the enterprise communication group, and handle it in real time in the communication group.
[0052] Although both of the above-mentioned two processing methods can handle exceptions, the exception reporting and processing paths of these two methods are separated from the operating system, which makes the efficiency of exception reporting low and has a learning time cost, making it difficult to accurately and efficiently handle the exceptions of the operating system.
[0053] In order to solve the above problems, the present invention proposes an exception handling method, please refer to Figure 1 , Figure 1 A flowchart of an embodiment of the exception handling method provided by the present invention. The exception handling method provided in this embodiment is applied to a system consisting of a terminal device, a network and a server, wherein the network is a medium for providing a communication link between the terminal device and the server, which may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.; the operating system on the terminal device may include a handheld device operating system (iPhone operating system, iOS system), Android system or other operating system, and the terminal device is connected to the server through the network to achieve interaction, thereby performing operations such as receiving or sending data, and specifically may be various electronic devices with a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, portable computers and desktop servers, etc. Figure 1 As shown, the method is applied to the operating system and specifically includes the following steps:
[0054] S101. In response to an abnormality reporting instruction, receive description information of an abnormal event input by an agent.
[0055] In this embodiment, the operating system of a financial insurance enterprise generally refers to a software system used by customer service representatives (agents). The system supports agents to complete daily work tasks, such as handling insurance consultation, quotation, insurance, claims, etc. The operating system can be run on, for example, desktop computers, laptops, mobile devices, cloud terminals, and Web browsers, so that agents can access the operating system to efficiently complete daily work. An abnormality reporting function entry is integrated into the operating system, and a corresponding abnormality reporting button is set on the front-end interface of the operating system. When an abnormality occurs in the use of the operating system by the agent, the abnormality reporting instruction can be triggered by clicking the abnormality reporting button, thereby achieving the purpose of one-click abnormality reporting through a unified functional entry, without the need for agents to report abnormalities across systems from other channels, which is more convenient for agent operation and reduces learning costs.
[0056] When the agent clicks the exception reporting button to trigger the exception reporting instruction, the agent can enter the description information of the current exception event in the preset text box. The description information may include, for example, what kind of abnormal phenomenon occurred in the system at a specific date and time, such as system crash, unresponsive operation, inability to log in, data loss, response delay, etc. The description information entered by the agent is received and recorded, thereby providing a reliable troubleshooting basis for efficient exception handling.
[0057] S102: Input the description information of the abnormal event into a trained abnormal prediction classification model to perform abnormal classification on the abnormal event.
[0058] After receiving the description information of the abnormal event, while storing the information, the description information is also input into the trained abnormal prediction classification model to perform semantic analysis on the description text and thus realize the prediction and classification of the abnormal event. The abnormal prediction classification model is specifically trained by a deep learning model based on the Transformer architecture. It uses the core idea of Transformer to process sequence data and perform classification prediction. The abnormal prediction classification model can be trained through the following steps: collect historical abnormal data and divide it into training data sets, validation sets and test sets. The historical abnormal data includes abnormal descriptions, types, impacts, processing records, etc.; then clean the data, remove irrelevant or missing information, and perform word segmentation and standardization on the text data; extract useful features from the abnormal data, such as abnormal type, abnormal level, impact range, etc. For text data, word embedding (such as Word2Ve c, BERT) into a numerical vector; serialize the anomaly data, for example, convert the anomaly description into a fixed-length sequence; add position encoding to the sequence data so that the Transformer can capture the position information in the sequence; design the Transformer model architecture, including the number of encoder layers, the number of attention heads, the hidden layer dimensions, etc.; train the model using the training data set, adjust the hyperparameters to obtain the best performance, and use cross-validation to evaluate the generalization ability of the model; use the validation set and the test set to evaluate the model's accuracy, recall rate, F1 score and other indicators; optimize the model based on the evaluation results, such as adjusting the model structure, using regularization techniques, etc.; if there are fewer abnormal data of a specific type, transfer learning or fine-tuning techniques can be used to improve the model's recognition ability of these types, and finally a trained anomaly prediction and classification model is obtained, and the trained model is deployed to the production environment for real-time or batch anomaly classification.
[0059] The predictive classification model using the Transformer architecture uses a self-attention mechanism to capture long-distance dependencies within the sequence, which is crucial for understanding contextual information. The model also uses a multi-head attention mechanism to learn the features of the sequence from different perspectives at the same time, enhancing the model's representation capabilities. While achieving one-click convenient exception reporting through a unified function entry, it can accurately predict and classify exceptions based on descriptive information, further improving the efficiency of exception handling.
[0060] S103: Matching is performed in an exception handling library according to the exception classification result, wherein the exception handling library stores resolved exceptions and corresponding solutions.
[0061] Based on the anomaly type of the current abnormal event output by the anomaly prediction classification model, the anomaly type is matched in the pre-set anomaly handling library. Based on the anomalies that occurred in history and the data of the anomaly solutions, the anomaly handling library stores the resolved anomalies and the corresponding solutions. When matching, the keywords in the anomaly classification results can be used to search in the anomaly handling library to find relevant solutions; or the solutions in the anomaly handling library can be classified and labeled according to different anomaly types, and the corresponding labels can be directly matched according to the anomaly classification results; or the anomaly classification results can be combined with the anomaly description to calculate the similarity between the anomaly classification results and the resolved anomalies in the anomaly handling library to find the most similar cases and solutions; or the nearest neighbor search algorithm can be used to find the most similar solutions in the anomaly handling library according to the feature vector of the anomaly, etc. Specifically, the corresponding matching method can be selected according to the structure of the anomaly handling library, the characteristics of the anomaly data, the required matching accuracy, etc., and this embodiment does not limit this.
[0062] S104: When a target anomaly is matched with a corresponding self-service solution, the self-service solution is displayed on the terminal interface of the agent.
[0063] When the corresponding target exception and the corresponding self-service solution are successfully matched in the exception handling library, the self-service solution is directly pushed to the agent's terminal to display the solution on the interface. For example, the summary or detailed information of the self-service solution can be displayed in a pop-up window on the agent interface; or a quick access link or list of the solution can be integrated in the sidebar of the agent interface; or an interactive wizard is generated and displayed based on the self-service solution to guide the agent through each step of the self-service solution, etc. After the agent reports the exception in the operating system with one click, the agent can find solutions to similar abnormal events through the deployed exception prediction and classification model, thereby guiding the agent to operate based on the steps in the self-service solution to eliminate the exception, helping the agent to solve the problem by himself, and when the match fails in the exception handling library, the backend operation and maintenance personnel are notified for manual intervention, which can effectively improve the efficiency of the operating system exception handling and reduce labor costs.
[0064] The exception handling method provided in this embodiment integrates exception reporting with the operating system. The agent can report the description information of the exception event in the operating system with one click, which is convenient for the agent operation and effectively saves the learning time cost. After semantic analysis of the reported exception description information, the exception event is automatically classified and a matching self-service solution is fed back, thereby improving the efficiency of exception handling in the operating system.
[0065] In one embodiment, Figure 2 As shown, after responding to the abnormality reporting instruction, the method further includes:
[0066] S201, displaying an exception recurrence prompt, wherein the exception recurrence prompt is used to prompt the agent to input an operation to reproduce the exception;
[0067] S202: After detecting that the agent starts to input an operation to reproduce the exception, record all DOM events when the exception occurs.
[0068] In this embodiment, after the agent clicks the exception reporting button, not only the description information of the abnormal event input by the agent is received, but also the abnormal recurrence prompt is displayed before or after the agent inputs the description information, for example, it can be displayed in a pop-up window, guiding information, etc. The abnormal recurrence prompt is used to prompt the agent to input the operation to reproduce the abnormality. After detecting that the agent starts to input the operation to reproduce the abnormality, all DOM events when the abnormality occurs are recorded. For example, when the abnormal event is unable to log in, the operation that the agent needs to input is to enter the user name and password and then click the login button, and wait for a certain period of time for the server to respond, so as to completely record all DOM events when the abnormal event occurs. The recorded events mainly include: the browser snapshot data at the time of recording the start, and the time, content, log information, backend request and other information of 21 major events such as onchange (content change), onclick (user single click), ondbclick (user double click), onkeydown (a keyboard key is pressed), onmousemove (mouse movement), etc. By recording all DOM events when an exception occurs, more accurate auxiliary information can be provided when the description information reported by the agent is not accurate enough. It can also provide complete and reliable exception record data when manual intervention is required to solve the problem, thereby improving the accuracy of exception handling.
[0069] Among them, recording DOM events can be achieved by adding event listeners to DOM elements using the addEventListener method, or using event objects (event parameters) to access detailed event information, such as time, target, etc.; or directly using event attributes on HTML elements, such as onclick, onload, etc.; or setting listeners on parent elements to capture events of child elements through event bubbling, etc. The specific settings can be more flexible according to needs, and this embodiment does not limit this.
[0070] Specifically, recording can be started after detecting that the agent closes the prompt or the prompt countdown is automatically closed, so that the agent can directly input the operation to reproduce the exception after closing the prompt, and then record all DOM events when the exception occurs, thereby improving the efficiency of exception reporting; or an exception reproduction button can be set. When it is detected that the agent clicks the exception reproduction button, it is considered that the agent starts to input the operation to reproduce the exception. At this time, all DOM events when the exception occurs are recorded, so that the agent can flexibly control the input time of the exception reproduction according to his own work schedule, thereby avoiding the automatic recording method from failing to record reliable and complete exception events due to reasons such as the temporary departure of the agent.
[0071] In one embodiment, Figure 3 As shown, the method also includes:
[0072] S301: When the target exception cannot be matched or the self-service solution is invalid, mark the exception event as an unresolved exception;
[0073] S302: Obtain all DOM events of the unresolved exception and push them to the operation and maintenance end, so as to replay the unresolved exception on the operation and maintenance end;
[0074] S303: Receive and display a manual solution sent by the operation and maintenance end, where the manual solution is input by the operation and maintenance personnel based on the playback of the unresolved exception.
[0075] In this embodiment, when the self-service cannot solve the current exception event, that is, the target exception and the corresponding self-service solution cannot be matched in the exception handling library, or even if the self-service solution is matched, the exception event cannot be solved, then the exception event is marked as an unresolved exception, and manual intervention is required for the self-service unresolved exceptions. At this time, all DOM events of the unresolved exceptions are first obtained and pushed to the operation and maintenance end, so that the unresolved exceptions can be replayed at the operation and maintenance end. That is to say, when no solution is matched or the agent does not agree with the solution automatically fed back by the system and cannot solve the exception, the background staff will intervene to solve the problem. The background operation and maintenance personnel will pull the DOM events corresponding to the unresolved exceptions from the database, and the operation and maintenance end will use all DOM events to replay the agent's operations. In this way, the background operation and maintenance personnel can make a comprehensive assessment through information such as operation process, back-end requests, console logs, etc., and give a manual solution. The operation and maintenance end will push the received manual solution to the agent's terminal, and the terminal will receive and display the manual solution sent by the operation and maintenance end. Since the context of the abnormal operation and DOM events can be seen through the recorded screen playback, it is beneficial for the background operation and maintenance personnel to analyze and troubleshoot the problem. There is no need for the operation and maintenance personnel to remotely go to the agent's terminal to reproduce the problem and obtain relevant phenomenon logs, which effectively reduces the communication with the agent and improves work efficiency.
[0076] In one embodiment, Figure 4 As shown, the method also includes:
[0077] S401, collect statistics on abnormal information within a specified time period, wherein the abnormal information includes abnormal type, abnormal scope, and abnormal resolution time of the abnormal event;
[0078] S402: Determine the processing priorities of different exception types according to the statistical results of the exception information.
[0079] In this embodiment, since the agent reports abnormal problems online through a unified function portal, it is very convenient and quick to count historical abnormal events. For example, the abnormal information in the waiting time period within the past three months, six months, and one year can be counted, including the abnormal type, abnormal scope, abnormal resolution time, etc. of the abnormal event. Based on the statistical results of the abnormal information, the number, frequency of occurrence, scope of impact, resolution time, and other factors of various types of events can be analyzed, and then the processing priority of different abnormal types can be confirmed. For example, the frequency of occurrence of different abnormal types is analyzed, and frequently occurring abnormalities have a higher processing priority; different levels of severity (such as minor, medium, and severe) are defined according to the impact of the abnormality on the business, and the higher the severity, the higher the processing priority; the average time required to resolve different abnormalities is analyzed, and the abnormalities with a long resolution time may need to be processed first, so the processing priority is higher, etc., and each statistical result can be further assigned a weight, and the processing priority of different abnormal types is obtained after the comprehensive score of each abnormal type is calculated based on the statistical results, so that the priority result comprehensively considers different factors and is more in line with actual processing needs.
[0080] In one embodiment, Figure 5 As shown, when the target exception cannot be matched or the self-service solution is invalid, after marking the exception event as an unresolved exception, the method further includes:
[0081] S501, confirming the corresponding processing priority according to the exception classification result of each unresolved exception;
[0082] S502: Generate a dynamic exception resolution plan based on the processing priority of all unresolved exceptions and push it to the operation and maintenance end.
[0083] In this embodiment, based on the processing priority obtained by statistical analysis, when manual intervention is required to handle the exception, since there is usually more than one unresolved exception, in order to achieve a more reasonable and efficient work arrangement, the corresponding processing priority can be first confirmed according to the exception classification result of each unresolved exception, and a dynamic exception resolution plan can be generated according to the processing priority of all unresolved exceptions and pushed to the operation and maintenance end. The higher the priority of the unresolved exception in the exception resolution plan, the earlier the planned time is, and vice versa, the lower the priority of the unresolved exception, the later the actual plan is. The plan is dynamically updated based on the latest marked unresolved exceptions and synchronized to the operation and maintenance end, for example, it is updated once every time one or a fixed number of unresolved exceptions are marked, or it can also be updated once at regular intervals, so that the operation and maintenance end can receive the latest exception resolution plan. When manual intervention is performed to handle the exception, the exception resolution plan generated based on the processing priority is used as a reference for its work arrangement, and more important or urgent abnormal events are processed first, and the exception processing order is arranged more reasonably to avoid the impact of important or urgent abnormal events on business operations, customer satisfaction, system stability, etc. due to the fact that important or urgent abnormal events have not been resolved, thereby improving the processing efficiency and rationality of exception processing.
[0084] In one embodiment, the method further comprises:
[0085] Receive exception resolution results from agents.
[0086] In this embodiment, when the exception is resolved self-service or manually, a feedback information input page can be displayed on the agent's terminal interface to receive the exception resolution results fed back by the agent, which may include, for example, whether the exception is resolved, whether the exception resolution channel is self-service or manual, the exception resolution time, satisfaction with the self-service solution, etc. By recording and analyzing the exception resolution results fed back by the agent, it is beneficial to continuously improve the exception handling process and improve the exception handling efficiency and processing effect.
[0087] In one embodiment, Figure 6 As shown, after receiving the abnormality resolution result fed back by the agent, the method further includes:
[0088] confirming the matching degree between the abnormal classification result and the abnormal event according to the abnormal resolution result;
[0089] Feedback correction is performed on the abnormal prediction classification model according to the matching degree.
[0090] In this embodiment, the degree of match between the abnormal classification results output by the abnormal prediction classification model and the actual abnormal events is confirmed based on the abnormal resolution results fed back by the seats, for example, whether the abnormality is self-resolved, if so, the matching degree is high, and if it is solved by manual intervention, the matching degree is low; the abnormal resolution time is judged, the shorter the resolution time, the higher the matching degree; the higher the satisfaction with the self-service solution, the higher the matching degree, and so on. A mapping table can be pre-set based on these factors, and a variety of factors and corresponding quantitative matching scores are stored in the table. The matching score of each factor is obtained based on the distribution of the abnormal resolution results in the mapping table, and then the matching degree is calculated based on the weight of the factor, so as to match and quantify the abnormal resolution results fed back by the seats, and the abnormal prediction classification model is fed back and corrected with a more quantitative matching degree, such as parameter adjustment and other processing, so that the abnormal prediction classification model can be adaptively feedback-adjusted according to the actual abnormal processing results, so that the abnormal prediction classification model can be more in line with the abnormal conditions in the actual production environment, and the accuracy of abnormal prediction classification and the success rate of abnormal self-service resolution are improved.
[0091] It should be noted that there is not necessarily a certain order between the above-mentioned steps. A person skilled in the art can understand, based on the description of the embodiments of the present invention, that in different embodiments, the above-mentioned steps may have different execution orders, that is, they may be executed in parallel, may be executed interchangeably, and so on.
[0092] Another embodiment of the present invention provides an exception handling device, such as Figure 7 As shown, the device 1 comprises:
[0093] The abnormality reporting module 11 is used to receive the description information of the abnormal event input by the agent in response to the abnormality reporting instruction;
[0094] A prediction and classification module 12 is used to input the description information of the abnormal event into the trained abnormal prediction and classification model to classify the abnormal event;
[0095] An exception matching module 13 is used to match in an exception handling library according to the exception classification result, wherein the exception handling library stores resolved exceptions and corresponding solutions;
[0096] The solution display module 14 is used to display the self-service solution on the terminal interface of the agent when a target anomaly is matched with a corresponding self-service solution.
[0097] The module referred to in the present invention refers to a series of computer program instruction segments that can perform specific functions, which are more suitable for describing the exception handling execution process than programs. For the specific implementation of each module, please refer to the corresponding method embodiment above, which will not be repeated here.
[0098] In one embodiment, the device 1 further comprises:
[0099] A prompt module, used to display an exception recurrence prompt, wherein the exception recurrence prompt is used to prompt the agent to input an operation to reproduce the exception;
[0100] The recording module is used to record all DOM events when the exception occurs after detecting that the agent starts to input the operation to reproduce the exception.
[0101] In one embodiment, the device 1 further comprises:
[0102] A marking module, used for marking the abnormal event as an unresolved abnormality when no target abnormality is matched or the self-service solution is invalid;
[0103] An acquisition push module is used to acquire all DOM events of the unresolved exceptions and push them to the operation and maintenance end, so as to replay the unresolved exceptions at the operation and maintenance end;
[0104] The solution receiving module is used to receive and display the manual solution sent by the operation and maintenance end, where the manual solution is input by the operation and maintenance personnel based on the playback of the unresolved exception.
[0105] In one embodiment, the device 1 further comprises:
[0106] A statistics module is used to collect statistics on abnormal information within a specified time period, wherein the abnormal information includes the abnormal type, abnormal range, and abnormal resolution time of the abnormal event;
[0107] The priority module is used to determine the processing priority of different exception types according to the statistical results of the exception information.
[0108] In one embodiment, the device 1 further comprises:
[0109] A confirmation module is used to confirm the corresponding processing priority according to the exception classification result of each unresolved exception;
[0110] The plan push module is used to generate a dynamic exception resolution plan based on the processing priority of all unresolved exceptions and push it to the operation and maintenance end.
[0111] In one embodiment, the device 1 further comprises:
[0112] The feedback module is used to receive the exception resolution results fed back by the agents.
[0113] In one embodiment, the device 1 further comprises:
[0114] A matching module, used to confirm the matching degree between the abnormal classification result and the abnormal event according to the abnormal resolution result;
[0115] A model correction module is used to perform feedback correction on the abnormality prediction classification model according to the matching degree.
[0116] Another embodiment of the present invention provides an exception handling system, such as Figure 8 As shown, the system 10 includes:
[0117] One or more processors 110 and memory 120, Figure 8 A processor 110 is used as an example for description. The processor 110 and the memory 120 may be connected via a bus or other means. Figure 8 The example of connecting through bus is taken in the following.
[0118] The processor 110 is used to complete various control logics of the system 10, and it can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a single-chip microcomputer, an ARM (Acorn RISC Machine) or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination of these components. In addition, the processor 110 can also be any traditional processor, microprocessor or state machine. The processor 110 can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP and / or any other such configuration.
[0119] The memory 120 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions corresponding to the exception handling method in the embodiment of the present invention. The processor 110 executes various functional applications and data processing of the system 10 by running the non-volatile software programs, instructions and units stored in the memory 120, that is, implements the exception handling method in the above method embodiment.
[0120] The memory 120 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created according to the use of the system 10, etc. In addition, the memory 120 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 120 may optionally include a memory remotely arranged relative to the processor 110, and these remote memories may be connected to the system 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0121] One or more units are stored in the memory 120, and when executed by one or more processors 110, the exception handling method in any of the above method embodiments is executed, for example, the following steps are executed:
[0122] Applied to an operating system, the method comprises:
[0123] In response to the abnormality reporting instruction, receiving description information of the abnormal event input by the agent;
[0124] Inputting the description information of the abnormal event into the trained abnormal prediction classification model to perform abnormal classification on the abnormal event;
[0125] Matching is performed in an exception handling library according to the exception classification result, wherein the exception handling library stores resolved exceptions and corresponding solutions;
[0126] When the target anomaly is matched with the corresponding self-service solution, the self-service solution is displayed on the agent's terminal interface.
[0127] In one embodiment, after responding to the abnormality reporting instruction, the method further includes:
[0128] Displaying an exception reproduction prompt, wherein the exception reproduction prompt is used to prompt the agent to input an operation to reproduce the exception;
[0129] After detecting that the agent starts to input the operation to reproduce the exception, all DOM events when the exception occurs are recorded.
[0130] In one embodiment, the method further comprises:
[0131] When the target exception cannot be matched or the self-service solution is invalid, the exception event is marked as an unresolved exception;
[0132] Obtain all DOM events of the unresolved exception and push them to the operation and maintenance end, so as to replay the unresolved exception at the operation and maintenance end;
[0133] Receive and display a manual solution sent by the operation and maintenance end, where the manual solution is input by the operation and maintenance personnel based on the playback of the unresolved exception.
[0134] In one embodiment, the method further comprises:
[0135] Collect statistics on abnormal information within a specified time period, including the abnormal type, abnormal scope, and abnormal resolution time of the abnormal event;
[0136] The processing priorities of different exception types are determined based on the statistical results of the exception information.
[0137] In one embodiment, when the target exception cannot be matched or the self-service solution is invalid, after marking the exception event as an unresolved exception, the method further includes:
[0138] Determine the corresponding processing priority based on the exception classification results of each unresolved exception;
[0139] Generate a dynamic exception resolution plan based on the processing priority of all unresolved exceptions and push it to the operation and maintenance end.
[0140] In one embodiment, the method further comprises:
[0141] Receive exception resolution results from agents.
[0142] In one embodiment, after receiving the abnormality resolution result fed back by the agent, the method further includes:
[0143] confirming the matching degree between the abnormal classification result and the abnormal event according to the abnormal resolution result;
[0144] Feedback correction is performed on the abnormal prediction classification model according to the matching degree.
[0145] An embodiment of the present invention provides a non-volatile computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed by one or more processors, the exception handling method in any of the above method embodiments is executed, for example, the following steps are executed:
[0146] In response to the abnormality reporting instruction, receiving description information of the abnormal event input by the agent;
[0147] Inputting the description information of the abnormal event into the trained abnormal prediction classification model to perform abnormal classification on the abnormal event;
[0148] Matching is performed in an exception handling library according to the exception classification result, wherein the exception handling library stores resolved exceptions and corresponding solutions;
[0149] When the target anomaly is matched with the corresponding self-service solution, the self-service solution is displayed on the agent's terminal interface.
[0150] In one embodiment, after responding to the abnormality reporting instruction, the method further includes:
[0151] Displaying an exception reproduction prompt, wherein the exception reproduction prompt is used to prompt the agent to input an operation to reproduce the exception;
[0152] After detecting that the agent starts to input the operation to reproduce the exception, all DOM events when the exception occurs are recorded.
[0153] In one embodiment, the method further comprises:
[0154] When the target exception cannot be matched or the self-service solution is invalid, the exception event is marked as an unresolved exception;
[0155] Obtain all DOM events of the unresolved exception and push them to the operation and maintenance end, so as to replay the unresolved exception at the operation and maintenance end;
[0156] Receive and display a manual solution sent by the operation and maintenance end, where the manual solution is input by the operation and maintenance personnel based on the playback of the unresolved exception.
[0157] In one embodiment, the method further comprises:
[0158] Collect statistics on abnormal information within a specified time period, including the abnormal type, abnormal scope, and abnormal resolution time of the abnormal event;
[0159] The processing priorities of different exception types are determined based on the statistical results of the exception information.
[0160] In one embodiment, when the target exception cannot be matched or the self-service solution is invalid, after marking the exception event as an unresolved exception, the method further includes:
[0161] Determine the corresponding processing priority based on the exception classification results of each unresolved exception;
[0162] Generate a dynamic exception resolution plan based on the processing priority of all unresolved exceptions and push it to the operation and maintenance end.
[0163] In one embodiment, the method further comprises:
[0164] Receive exception resolution results from agents.
[0165] In one embodiment, after receiving the abnormality resolution result fed back by the agent, the method further includes:
[0166] confirming the matching degree between the abnormal classification result and the abnormal event according to the abnormal resolution result;
[0167] Feedback correction is performed on the abnormal prediction classification model according to the matching degree.
[0168] As an example, non-volatile storage media can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) as external cache memory. By way of illustration and not limitation, RAM can be obtained in many forms such as synchronous RAM (SRAM), dynamic RAM, (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The disclosed memory components or memories of the operating environment described herein are intended to include one or more of these and / or any other suitable types of memory.
[0169] In summary, the present invention discloses an exception handling method, device, system and medium, the method is applied to the operating system, including: in response to an exception reporting instruction, receiving the description information of the abnormal event input by the agent; inputting the description information of the abnormal event into the trained abnormal prediction classification model to classify the abnormal event; matching in the exception handling library according to the abnormal classification result, the exception handling library stores the resolved exceptions and corresponding solutions; when the target exception is matched with the corresponding self-service solution, the self-service solution is displayed on the terminal interface of the agent. By integrating the exception reporting with the operating system, the agent can report the description information of the abnormal event in the operating system with one click, which is convenient for the agent to operate and effectively saves the learning time cost. After the semantic analysis of the reported abnormal description information, the abnormal event is automatically classified and the matching self-service solution is fed back, thereby improving the efficiency of the operating system's exception handling.
[0170] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium, and the computer program can include the processes of the above-mentioned method embodiments when executed. The storage medium can be a memory, a disk, a floppy disk, a flash memory, an optical storage device, etc.
[0171] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.
Claims
1. An exception handling method, characterized in that: Applied to an operating system, the method comprises: In response to the abnormality reporting instruction, receiving description information of the abnormal event input by the agent; Inputting the description information of the abnormal event into the trained abnormal prediction classification model to perform abnormal classification on the abnormal event; Matching is performed in an exception handling library according to the exception classification result, wherein the exception handling library stores resolved exceptions and corresponding solutions; When the target anomaly is matched with the corresponding self-service solution, the self-service solution is displayed on the agent's terminal interface.
2. The exception handling method according to claim 1, characterized in that: After responding to the abnormality reporting instruction, the method further includes: Displaying an exception reproduction prompt, wherein the exception reproduction prompt is used to prompt the agent to input an operation to reproduce the exception; After detecting that the agent starts to input the operation to reproduce the exception, all DOM events when the exception occurs are recorded.
3. The exception handling method according to claim 2, characterized in that: The method further comprises: When the target exception cannot be matched or the self-service solution is invalid, the exception event is marked as an unresolved exception; Obtain all DOM events of the unresolved exception and push them to the operation and maintenance end, so as to replay the unresolved exception at the operation and maintenance end; Receive and display a manual solution sent by the operation and maintenance end, where the manual solution is input by the operation and maintenance personnel based on the playback of the unresolved exception.
4. The exception handling method according to claim 3, characterized in that: The method further comprises: Collect statistics on abnormal information within a specified time period, including the abnormal type, abnormal scope, and abnormal resolution time of the abnormal event; The processing priorities of different exception types are determined based on the statistical results of the exception information.
5. The exception handling method according to claim 4, characterized in that: When the target exception cannot be matched or the self-service solution is invalid, after marking the exception event as an unresolved exception, the method further includes: Determine the corresponding processing priority based on the exception classification results of each unresolved exception; Generate a dynamic exception resolution plan based on the processing priority of all unresolved exceptions and push it to the operation and maintenance end.
6. The exception handling method according to any one of claims 1 to 5, characterized in that: The method further comprises: Receive exception resolution results from agents.
7. The exception handling method according to claim 6, characterized in that: After receiving the abnormality resolution result fed back by the agent, the method further includes: confirming the matching degree between the abnormal classification result and the abnormal event according to the abnormal resolution result; Feedback correction is performed on the abnormal prediction classification model according to the matching degree.
8. An exception handling device, characterized in that: include: The abnormality reporting module is used to receive the description information of the abnormal event input by the seat in response to the abnormality reporting instruction; A prediction and classification module, used to input the description information of the abnormal event into the trained abnormal prediction and classification model to perform abnormal classification on the abnormal event; An exception matching module, used to match in an exception handling library according to the exception classification result, wherein the exception handling library stores resolved exceptions and corresponding solutions; The solution display module is used to display the self-service solution on the terminal interface of the agent when the target exception is matched with the corresponding self-service solution.
9. An exception handling system, characterized in that: The system includes at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the exception handling method according to any one of claims 1 to 7.
10. A non-volatile computer-readable storage medium, characterized in that: The non-volatile computer-readable storage medium stores computer-executable instructions, which, when executed by one or more processors, enable the one or more processors to execute the exception handling method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Processing method and terminal of client feedback information
CN107833056A
Abnormity processing method and device, electronic equipment and computer readable storage medium
CN112182149A
Abnormal program processing method and system
CN117421147A
Probabilistic anomaly detection in streaming device data
US20220114593A1
Cited By
Intelligent processing method, device and equipment for transaction abnormity and storage medium
CN119963185A
