Atomic service update method, apparatus, device, and medium

Abstract

This invention discloses an atomic service update method, apparatus, device, and medium. The method includes: acquiring user satisfaction feedback data and service operation data of the atomic service; determining a comprehensive satisfaction score for the atomic service based on the data type of the satisfaction feedback data; determining a target update strategy for the atomic service based on the comprehensive satisfaction score and service operation data; and updating the atomic service based on the target update strategy. This method identifies and quantifies the feedback data to determine the comprehensive satisfaction score and, combined with the service operation data, generates a target update strategy, thereby achieving autonomous optimization of the atomic service, improving service stability and optimization efficiency, and contributing to a better user experience.

Description

Technical Field

[0001] This invention relates to the field of intelligent service technology, and in particular to an atomic service update method, apparatus, device, and medium. Background Technology

[0002] As enterprise services evolve towards greater intelligence, systems are shifting from static question-and-answer mechanisms to process-driven task handling, such as automated expense reimbursement, attendance correction, and data entry. However, existing atomic services are typically predefined functionalities whose service logic cannot automatically adjust to business evolution, nor can they perceive the actual user experience after invoking them. Furthermore, they cannot promptly and adaptively optimize atomic services when user experience is poor. Therefore, a method is urgently needed to optimize atomic services in a timely manner based on user experience. Summary of the Invention

[0003] This invention provides an atomic service update method, apparatus, device, and medium to address the problem of untimely service updates caused by a lack of user experience perception in atomic services in related technologies.

[0004] In a first aspect, the present invention provides an atomic service update method, comprising: Obtain user satisfaction feedback data and service operation data for atomic services; Based on the data type of the satisfaction feedback data, determine the comprehensive satisfaction score of the atomic service; Based on comprehensive satisfaction scores and service operation data, determine the target update strategy for atomic services: The atomic services are updated based on the target update strategy.

[0005] In some embodiments, obtaining user satisfaction feedback data for atomic services includes at least one of the following: Based on the application programming interface (API), obtain user-uploaded satisfaction feedback data; Based on the service exception status information in the call logs of atomic services, obtain satisfaction feedback data; Obtain user feedback information from the upper-layer service of the atomic service, and obtain satisfaction feedback data.

[0006] In some embodiments, based on the data type of the satisfaction feedback data, a comprehensive satisfaction score for the atomic service is determined, including: Type identification is performed on satisfaction feedback data to determine structured feedback data, unstructured feedback data, and log data; The structured feedback data is quantified, and the results of the quantification are weighted and fused to obtain a structured satisfaction score. Feature extraction is performed on unstructured feedback data, and the results of feature extraction are quantified to obtain unstructured satisfaction scores. Based on a preset log rating mapping relationship, the log data is rated and mapped to obtain a log satisfaction rating; A comprehensive satisfaction score is determined based on structured satisfaction scores, unstructured satisfaction scores, and log satisfaction scores.

[0007] In some embodiments, a target update strategy for atomic services is determined based on comprehensive satisfaction scores and service operation data, including: Perform anomaly identification on service operation data to determine abnormal service parameters and duration; Based on service anomaly parameters and service anomaly duration, identify the anomaly factors in service operation data; Attribution analysis was conducted on the abnormal factors to determine their impact on the overall satisfaction score. Based on a comprehensive score of abnormal factors, degree of impact, and satisfaction, a target update strategy is determined.

[0008] In some embodiments, attribution analysis is performed on the abnormal factors to determine the degree of influence of the abnormal factors on the overall satisfaction score, including: Obtain the historical mapping relationship between abnormal factors and overall satisfaction scores; Based on historical mapping relationships, determine the update priority of abnormal factors; The degree of impact is determined based on update priority.

[0009] In some embodiments, a target update strategy is determined based on a comprehensive score of anomalies, impact level, and satisfaction, including: Based on the abnormal factors, multiple candidate update strategies are generated; Based on the comprehensive score of impact and satisfaction, simulation verification is carried out on each candidate update strategy to simulate the satisfaction improvement effect of multiple candidate update strategies. Based on the effect of improving satisfaction, the target update strategy is determined from multiple candidate update strategies.

[0010] In some embodiments, after determining the overall satisfaction score for the atomic service, the method further includes: Retrieve the service call result and service call target of the atomic service; When the service call result is a service call failure and / or the overall satisfaction score is lower than the preset score threshold, the atomic service is interrupted, and the service call goal is achieved based on the preset intelligent agent.

[0011] Secondly, this disclosure provides an atomic service update apparatus, comprising: include: The data acquisition module is used to acquire user satisfaction feedback data and service operation data for atomic services; The satisfaction rating module is used to determine the overall satisfaction rating of atomic services based on the data type of satisfaction feedback data. The strategy update module is used to determine the target update strategy for atomic services based on the overall satisfaction score and service operation data. The service update module is used to update atomic services based on the target update strategy.

[0012] Thirdly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described atomic service update method.

[0013] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described atomic service update method.

[0014] The aforementioned atomic service update method, apparatus, device, and medium implement a solution whereby the method includes: acquiring user satisfaction feedback data and service operation data for the atomic service; determining a comprehensive satisfaction score for the atomic service based on the data type of the satisfaction feedback data; determining a target update strategy for the atomic service based on the comprehensive satisfaction score and service operation data; and updating the atomic service based on the target update strategy. This method identifies and quantifies the feedback data to determine the comprehensive satisfaction score and, combined with service operation data, generates a target update strategy, thereby achieving autonomous optimization of the atomic service, improving service stability and optimization efficiency, and contributing to a better user experience. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of an atomic service update method in one embodiment of the present invention; Figure 2 This is a flowchart of an atomic service update method in one embodiment of the present invention; Figure 3 This is a flowchart of an atomic service update method in one embodiment of the present invention; Figure 4This is a flowchart of an atomic service update method in one embodiment of the present invention; Figure 5 This is a schematic block diagram of an atomic service update device according to an embodiment of the present invention; Figure 6 This is a schematic block diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0017] In one embodiment, such as Figure 1 As shown, an atomic service update method is provided, including: S101, Obtain user satisfaction feedback data and service operation data for atomic services; S102, Determine the overall satisfaction score of the atomic service based on the data type of the satisfaction feedback data; S103, Based on the comprehensive satisfaction score and service operation data, determine the target update strategy for atomic services: S104, Update the atomic service based on the target update strategy.

[0018] As an example, in step S101, structured or unstructured feedback data such as ratings and comments actively submitted by users can be obtained through the Application Programming Interface (API) to directly obtain user satisfaction feedback data on the atomic service; or user satisfaction feedback data can be indirectly obtained by analyzing information such as exception status codes, response delays and error stacks in the log files corresponding to the atomic service, but this disclosure does not limit the method of obtaining satisfaction feedback data.

[0019] In one embodiment, the performance indicators of the atomic service during operation can also be determined by analyzing the log data of the atomic service, thereby obtaining service operation data. However, this is not limited to this, and the present disclosure does not limit the method of obtaining service operation data.

[0020] It should be understood that satisfaction feedback data can be structured data (such as explicit ratings and option selections), unstructured data (such as user comments and voice feedback), or semi-structured log data (such as HTTP status codes and timeout markers); service operation data can cover real-time performance parameters such as response time, throughput, and memory utilization, as well as parameters that can be used to indicate the operating status of atomic services, such as the success rate and latency of downstream dependent services. This disclosure does not limit the data formats included in satisfaction feedback data and service operation data.

[0021] In one embodiment, satisfaction feedback data and service operation data can be acquired periodically, or collected in real time when events such as service anomalies, a surge in user complaints, or a sharp drop in ratings are detected, or acquired randomly within a preset time interval. This disclosure does not limit the timing of acquiring the above data.

[0022] As an example, in step S102, the satisfaction feedback data can be type-identified to distinguish different types of satisfaction feedback data, so that different scoring methods can be used for different types of satisfaction feedback data, and the scored data can be weighted and fused to generate a comprehensive satisfaction score.

[0023] When the satisfaction feedback data contains only one type of data, the score of that type of data can be directly used as the comprehensive satisfaction score; when it contains multiple data types, the scores of different data types are calculated separately and then weighted and merged to obtain the comprehensive satisfaction score.

[0024] In one embodiment, the data types can be divided into three categories: structured feedback data, unstructured feedback data, and log data, in order to achieve comprehensive coverage of satisfaction feedback data. However, it is not limited to this. Data types can also be divided into other categories, or the above data types can be further subdivided, as long as it can be ensured that the satisfaction feedback data of each data type is effectively identified and quantified.

[0025] As an example, in step S103, the preset satisfaction threshold can be determined based on the determined comprehensive satisfaction score, and the atomic service operation status can be jointly determined by combining the abnormal data in the service operation data, thereby determining the target update strategy for the atomic service.

[0026] In one embodiment, there may be only one satisfaction threshold. When the overall satisfaction score is lower than this threshold, the atomic service is considered to need to be updated. At this time, the abnormal factors that cause the low overall satisfaction score can be analyzed by combining service operation data, and a corresponding target update strategy can be generated based on this. When there are multiple satisfaction thresholds, the target update strategy with different granularities is matched according to the range of the overall satisfaction score. For example, if the score is lower than the first satisfaction threshold but not lower than the second satisfaction threshold, the target update strategy is determined to be parameter fine-tuning. If the score is lower than the second satisfaction threshold, the target update strategy is determined to be service reconstruction, etc.

[0027] It should be understood that when the overall satisfaction score meets the preset score requirements, it can be determined that the atomic service is running stably and the user satisfaction is high. At this time, the target update strategy can be determined to be empty.

[0028] In one embodiment, a corresponding target update strategy can be automatically generated based on the joint analysis results of the comprehensive satisfaction score and service status data, according to a preset rule engine and machine learning model. However, this is not limited to this, and the specific generation method of the target update strategy is not limited in this disclosure.

[0029] As an example, in step S104, atomic service update operations can be performed according to the target update strategy. That is, according to the specific operation instructions of the target update strategy, the service configuration interface is called to complete parameter adjustment, version upgrade or dependency replacement, so as to achieve timely optimization of atomic services.

[0030] In summary, this disclosure proposes an atomic service update method, comprising: obtaining user satisfaction feedback data and service operation data of the atomic service; determining a comprehensive satisfaction score for the atomic service based on the data type of the satisfaction feedback data; determining a target update strategy for the atomic service based on the comprehensive satisfaction score and service operation data; and updating the atomic service based on the target update strategy. This method identifies and quantifies the feedback data to determine the comprehensive satisfaction score, and combines this with service operation data to generate a target update strategy, thereby achieving autonomous optimization of the atomic service, improving service stability and optimization efficiency, and contributing to a better user experience.

[0031] In one embodiment, step S101, namely obtaining user satisfaction feedback data for the atomic service, includes at least one of the following: obtaining user-uploaded satisfaction feedback data based on the application programming interface; obtaining satisfaction feedback data based on service exception status information in the call log of the atomic service; obtaining user feedback information from the upper-layer service of the atomic service to obtain satisfaction feedback data.

[0032] In other words, satisfaction feedback data can be collected from multiple sources, including user-initiated feedback, log-based feedback, and feedback from upper-level services, thus avoiding evaluation distortion caused by biases from a single data source.

[0033] Specifically, user-submitted satisfaction ratings, text comments, or feature suggestions can be obtained through API interfaces. Alternatively, statistical analysis of abnormal states (such as timeouts, retries, and error codes) and sudden changes in call frequency in the call logs can be used to determine the results as satisfaction feedback data. Furthermore, user satisfaction data collected by upper-layer services of the atomic service can be used as satisfaction feedback data strongly correlated with that atomic service. These upper-layer services, such as service components or service systems containing the atomic service, are not limited in this disclosure. Because of their strong correlation with the atomic service, user feedback on the upper-layer service can indirectly reflect the user's evaluation of the atomic service; therefore, this feedback information can be determined as the satisfaction feedback data for that atomic service. This feedback information can include, for example, user ratings of the overall experience of the upper-layer service and optimization suggestions.

[0034] In one embodiment, such as Figure 2 As shown, step S102, which is to determine the comprehensive satisfaction score of the atomic service based on the data type of the satisfaction feedback data, includes: S201, Identify the type of satisfaction feedback data to determine structured feedback data, unstructured feedback data, and log data; S202, quantify the structured feedback data, and then weight and fuse the results of the quantification to obtain a structured satisfaction score; S203, extract features from unstructured feedback data and quantify the results of feature extraction to obtain unstructured satisfaction scores; S204. Based on the preset log rating mapping relationship, the log data is rated and mapped to obtain the log satisfaction rating. S205. Determine the comprehensive satisfaction score based on structured satisfaction scores, unstructured satisfaction scores, and log satisfaction scores.

[0035] As an example, in step S201, the type to which the satisfaction feedback data belongs can be determined by identifying and analyzing the data format, field structure, etc.

[0036] For example, it can identify JSON-formatted rating data as structured feedback data, plain text comments as unstructured feedback data, and log entries containing fields such as timestamps, error codes, and response times as log data.

[0037] As an example, in step S202, since the structured feedback data has clear fields and values, it can be directly mapped to the scores of the corresponding dimensions, thereby realizing the quantitative processing of the structured feedback data. The structured satisfaction score is generated by weighting and fusing the scores obtained after quantitative processing. The weight of each quantitative processing result can be dynamically adjusted according to the degree of influence of each field on user experience. For example, the weight of response time can be higher than the weight of the number of calls. This disclosure does not limit this.

[0038] For example, taking structured feedback data containing three fields: response time, error rate, and call success rate, these can be mapped to values ​​in the range of 0–10 (for example, 10 points for a response time ≤ 200ms, and 1 point deducted for every 100ms increase), and then weighted and summed according to preset weights (for example, 0.4, 0.3, and 0.3 respectively) to obtain a structured satisfaction score.

[0039] As an example, in step S203, since unstructured feedback data lacks a unified format, feature extraction is required based on its specific type. For example, semantic extraction models are used to identify semantic features such as emotional tendencies, key demands, and problem categories for text comments; speech recognition and voiceprint sentiment analysis are used to extract acoustic features such as emotional intensity and speech rate changes for audio messages; and OCR and visual semantic models are used to identify image features such as interface anomalies and layout errors for image feedback. Then, based on the extracted features, quantitative processing is performed to generate an unstructured satisfaction score.

[0040] For example, taking the three semantic features extracted as sentiment tendency, problem category and appeal intensity, the above three semantic features can be mapped to quantitative values ​​in the range of 0-10 (e.g., positive sentiment gets 8-10 points, neutral gets 5-7 points, and negative gets 0-4 points), and then weighted and fused according to preset weights (e.g., 0.5, 0.3, 0.2) to finally generate an unstructured satisfaction score.

[0041] As an example, in step S204, since the log data may be in unstructured or semi-structured form, it needs to be standardized and transformed according to the preset log scoring mapping relationship. For example, HTTP 500 errors are mapped to -3 points, 404 errors are mapped to -5 points, and response time less than 200ms is mapped to 4 points, so as to determine the quantitative result of each log data. Then, the log satisfaction score is obtained by weighted summation.

[0042] As an example, in step S205, structured satisfaction scores, unstructured satisfaction scores, and log satisfaction scores can be merged according to preset fusion weights to obtain the final comprehensive satisfaction score; alternatively, the fusion weights can be dynamically adjusted based on the amount of data and data confidence level of each type of satisfaction feedback data to more accurately reflect the user's real experience.

[0043] It should be understood that the weights used in steps S201-S205 can be preset or dynamically determined according to the actual situation, and this disclosure does not limit them.

[0044] In one embodiment, such as Figure 3 As shown, step S103, which is to determine the target update strategy for the atomic service based on the comprehensive satisfaction score and service operation data, includes: S301, identify anomalies in service operation data and determine abnormal service parameters and duration; S302, Based on service anomaly parameters and service anomaly duration, determine the anomaly factors of service operation data; S303, conduct attribution analysis on abnormal factors to determine the degree of influence of abnormal factors on the overall satisfaction score; S304. Based on the comprehensive score of abnormal factors, degree of impact, and satisfaction, determine the target update strategy.

[0045] As an example, in step S301, anomalies in service operation data can be identified by comparing key indicators such as response latency, error rate, and timeout count in the service operation data with thresholds; anomalies can also be identified by comparing the historical distribution of the service operation data with the historical distribution of the historical operation data, thereby achieving anomaly identification. However, this disclosure does not limit the specific method of anomaly identification.

[0046] The service operation data may include performance data (such as response time, throughput, resource utilization, etc.), dependency data (such as downstream service response time, interface call success rate), and environmental data (such as network latency, node load status), etc., which are not limited in this disclosure.

[0047] As an example, in step S302, the cause of the atomic service's anomaly is determined by combining the type and duration of the abnormal parameters, and the cause is identified as an abnormal factor.

[0048] For example, if the abnormal parameters indicate a sudden increase in response latency accompanied by a surge in CPU utilization, and the duration of the abnormality exceeding the threshold is determined based on the duration, then the abnormal factor can be identified as a resource bottleneck of the service node.

[0049] As an example, in step S303, since there may be more than one abnormal factor, and correcting all abnormal factors may lead to a waste of resources or excessive update time, it is necessary to perform attribution analysis on the abnormal factors to determine the degree of influence of each abnormal factor on the decline of the overall satisfaction score, so as to prioritize the factors with a greater degree of influence.

[0050] The degree of impact can be determined by examining the service scenarios corresponding to each abnormal factor. For example, in a query scenario, the impact of service node resource bottlenecks is significantly higher than that of network jitter. Alternatively, the actual impact of different abnormal factors on the overall satisfaction score in history can be determined by the historical mapping relationship between each abnormal factor and the overall satisfaction score, thereby determining the degree of impact.

[0051] In one embodiment, the historical mapping relationship between abnormal factors and comprehensive satisfaction scores can be obtained; based on the historical mapping relationship, the update priority of abnormal factors can be determined; based on the update priority, the degree of influence can be determined.

[0052] Specifically, by accessing historical databases or training models offline, the historical mapping relationship between abnormal factors and the overall satisfaction score can be extracted (for example, by obtaining data on the changing trends and magnitudes of the overall satisfaction score for each abnormal factor and the corresponding period within the past six months, thus obtaining the aforementioned historical mapping relationship); then, by performing statistical and regression analysis on this historical mapping relationship, the sensitivity of the overall satisfaction score to each abnormal factor can be determined, thereby determining the update priority of each abnormal factor; finally, based on the ranking of sensitivity, the degree of influence of each abnormal factor in the current overall satisfaction score can be determined.

[0053] It should be understood that the relationship between update priority and impact level can be dynamic. For example, when the impact levels of the identified anomalous factors are similar, the impact level can be determined by combining the frequency and duration of each influencing factor. For instance, anomalous factors with high frequency and long duration can be assigned a higher impact level.

[0054] As an example, in step S304, a comprehensive analysis can be conducted based on the abnormal factors, the degree of impact, and the current comprehensive satisfaction score to determine the corresponding target update strategy in order to improve the operational quality and comprehensive satisfaction score of the atomic service. The target update strategy can coordinately optimize all abnormal factors to achieve multi-dimensional synergistic improvement; or it can focus on intervening in the single abnormal factor with the highest degree of impact to improve update efficiency and resource utilization.

[0055] For example, when a service node resource bottleneck is identified as the most impactful abnormal factor, the target update strategy can be determined as dynamically expanding the CPU and memory quotas of the corresponding node.

[0056] In one embodiment, multiple candidate update strategies can be generated based on abnormal factors; each candidate update strategy can be simulated and verified based on the comprehensive score of influence and satisfaction, and the satisfaction improvement effect corresponding to multiple candidate update strategies can be simulated; based on the satisfaction improvement effect, the target update strategy can be determined from multiple candidate update strategies.

[0057] In other words, to avoid the target update strategy failing to achieve the expected results after application, simulation verification can be conducted to evaluate in advance the expected improvement of the overall satisfaction score and the impact on service operation stability of each candidate update strategy, thereby selecting the target update strategy.

[0058] Specifically, multiple candidate update strategies containing different update measures can be generated based on abnormal factors. Then, simulations can be performed based on the degree of impact and the comprehensive satisfaction score to simulate the changing trend of the comprehensive satisfaction score after the implementation of each candidate update strategy. This allows the satisfaction improvement effect corresponding to each candidate update strategy to be obtained. The candidate update strategy with the best satisfaction improvement effect can then be selected as the target update strategy. Alternatively, a comprehensive evaluation can be conducted by combining implementation cost, implementation cycle, and satisfaction improvement effect, and the candidate update strategy with the best comprehensive evaluation can be selected as the target update strategy. However, this is not limited to these methods. Other evaluation dimensions can also be set according to actual business needs, such as service availability, response latency, and resource consumption, to screen candidate update strategies and determine the target update strategy. This disclosure does not limit this approach.

[0059] In one embodiment, the candidate update strategy may include a collaborative optimization scheme for all abnormal factors, or it may include a targeted optimization scheme for only one or a few abnormal factors with a significant impact. At the same time, the update measures included in different candidate update strategies may differ in terms of technical implementation path, resource investment intensity, and effective time, so as to reflect the diversity of candidate update strategies and lay the foundation for ensuring the accuracy and adaptability of the target update strategy.

[0060] In one embodiment, when performing simulation verification on each candidate update strategy, the implementation records of update strategies that are similar to or the same as each candidate update strategy in history and the changes in the corresponding comprehensive satisfaction scores can be referenced to enhance the credibility and generalization ability of the simulation results.

[0061] In one embodiment, candidate update strategies can be generated using a preset rule base, a trained random forest model, a neural network model, etc., but are not limited to these.

[0062] In one embodiment, such as Figure 4 As shown, in step S102, after determining the overall satisfaction score of the atomic service, the following is also included: S401, retrieve the service call result and service call target of the atomic service; S402, when the service call result is a service call failure and / or the overall satisfaction score is lower than the preset score threshold, the atomic service is interrupted, and the service call target is achieved based on the preset intelligent agent.

[0063] As an example, in step S401, the service call result and service call target of the atomic service can be obtained through service call logs, monitoring metrics, or user feedback data. The service call result is used to indicate whether the user has successfully called the atomic service, while the service call target is used to indicate the specific purpose that the user expects to achieve by calling the atomic service.

[0064] As an example, in step S402, when the service call result is a service call failure and / or the overall satisfaction score is lower than the preset score threshold, it indicates that the atomic service cannot currently meet the user's needs or expected experience. At this time, the atomic service can be handed over to the preset intelligent agent for target takeover, and the intelligent agent will realize the service call target to ensure that business continuity and user experience are not interrupted.

[0065] It should be understood that the intelligent agent has independent decision-making and execution capabilities and can achieve service invocation goals based on historical takeover records and real-time information. This disclosure does not limit the specific implementation form, invocation interface and deployment method of the intelligent agent, nor does it limit the training method of the intelligent agent, as long as it is ensured that it has the corresponding functions.

[0066] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0067] In one embodiment, an atomic service update apparatus is provided, which corresponds one-to-one with the atomic service update method described in the above embodiments. For example... Figure 5 As shown, the atomic service update device includes a data acquisition module 501, a satisfaction rating module 502, a strategy update module 503, and a service update module 504. Detailed descriptions of each functional module are as follows: Data acquisition module 501 is used to acquire user satisfaction feedback data and service operation data for atomic services; The satisfaction rating module 502 is used to determine the overall satisfaction rating of the atomic service based on the data type of the satisfaction feedback data. The strategy update module 503 is used to determine the target update strategy for atomic services based on the comprehensive satisfaction score and service operation data. Service update module 504 is used to update atomic services based on the target update strategy.

[0068] In one embodiment, the data acquisition module 501 is further configured to acquire user-uploaded satisfaction feedback data based on the application programming interface; Based on the service exception status information in the call logs of atomic services, obtain satisfaction feedback data; Obtain user feedback information from the upper-layer service of the atomic service, and obtain satisfaction feedback data.

[0069] In one embodiment, the satisfaction rating module 502 is further configured to identify the type of satisfaction feedback data and determine structured feedback data, unstructured feedback data, and log data. The structured feedback data is quantified, and the results of the quantification are weighted and fused to obtain a structured satisfaction score. Feature extraction is performed on unstructured feedback data, and the results of feature extraction are quantified to obtain unstructured satisfaction scores. Based on a preset log rating mapping relationship, the log data is rated and mapped to obtain a log satisfaction rating; A comprehensive satisfaction score is determined based on structured satisfaction scores, unstructured satisfaction scores, and log satisfaction scores.

[0070] In one embodiment, the policy update module 503 is further configured to identify anomalies in service operation data and determine service anomaly parameters and service anomaly duration. Based on service anomaly parameters and service anomaly duration, identify the anomaly factors in service operation data; Attribution analysis was conducted on the abnormal factors to determine their impact on the overall satisfaction score. Based on a comprehensive score of abnormal factors, degree of impact, and satisfaction, a target update strategy is determined.

[0071] In one embodiment, the strategy update module 503 is further configured to perform attribution analysis on the abnormal factors to determine the degree of influence of the abnormal factors on the overall satisfaction score, including: Obtain the historical mapping relationship between abnormal factors and overall satisfaction scores; Based on historical mapping relationships, determine the update priority of abnormal factors; The degree of impact is determined based on update priority.

[0072] In one embodiment, the strategy update module 503 is further configured to determine a target update strategy based on a comprehensive score of abnormal factors, degree of impact, and satisfaction, including: Based on the abnormal factors, multiple candidate update strategies are generated; Based on the comprehensive score of impact and satisfaction, simulation verification is carried out on each candidate update strategy to simulate the satisfaction improvement effect of multiple candidate update strategies. Based on the effect of improving satisfaction, the target update strategy is determined from multiple candidate update strategies.

[0073] In one embodiment, the satisfaction rating module 502 is further used to obtain the service call result and service call target of the atomic service; When the service call result is a service call failure and / or the overall satisfaction score is lower than the preset score threshold, the atomic service is interrupted, and the service call goal is achieved based on the preset intelligent agent.

[0074] This invention provides an atomic service update device, comprising: a data acquisition module for acquiring user satisfaction feedback data and service operation data of the atomic service; a satisfaction scoring module for determining a comprehensive satisfaction score for the atomic service based on the data type of the satisfaction feedback data; a strategy update module for determining a target update strategy for the atomic service based on the comprehensive satisfaction score and service operation data; and a service update module for updating the atomic service based on the target update strategy. This device identifies and quantifies the feedback data to determine the comprehensive satisfaction score and, combined with the service operation data, generates a target update strategy, thereby achieving autonomous optimization of the atomic service, improving service stability and optimization efficiency, and contributing to a better user experience.

[0075] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database is used for data employed in the atomic service update method. The network interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements an atomic service update method.

[0076] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements an atomic service update method.

[0077] In one embodiment, a computer-readable storage medium is provided that stores a computer program, which, when executed by a processor, implements an atomic service update method.

[0078] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), IAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0079] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0080] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. An atomic service update method, characterized by, include: Obtain user satisfaction feedback data and service operation data for atomic services; Based on the data type of the satisfaction feedback data, determine the overall satisfaction score of the atomic service; Based on the overall satisfaction score and the service operation data, determine the target update strategy for the atomic service: The atomic service is updated based on the target update strategy.

2. The method of claim 1, wherein, The acquisition of user satisfaction feedback data for atomic services includes at least one of the following: Based on the application programming interface (API), obtain user-uploaded satisfaction feedback data; Based on the service exception status information in the call log of the atomic service, the satisfaction feedback data is obtained; Obtain user feedback information from the upper-layer service of the atomic service, and obtain the satisfaction feedback data.

3. The method of claim 1, wherein, The determination of the overall satisfaction score for the atomic service based on the data type of the satisfaction feedback data includes: The satisfaction feedback data is type-identified to determine structured feedback data, unstructured feedback data, and log data; The structured feedback data is quantified, and the results of the quantification are weighted and fused to obtain a structured satisfaction score. Feature extraction is performed on the unstructured feedback data, and the results of the feature extraction are quantified to obtain an unstructured satisfaction score. Based on a preset log rating mapping relationship, the log data is rated and mapped to obtain a log satisfaction rating; The overall satisfaction score is determined based on the structured satisfaction score, the unstructured satisfaction score, and the log satisfaction score.

4. The method of claim 1, wherein, The step of determining the target update strategy for the atomic service based on the overall satisfaction score and the service operation data includes: The service operation data is anomaly identified to determine the abnormal service parameters and duration. Based on the service anomaly parameters and the service anomaly duration, the abnormal factors of the service operation data are determined; Attribution analysis was performed on the abnormal factors to determine the degree of influence of the abnormal factors on the overall satisfaction score; The target update strategy is determined based on the abnormal factors, the degree of impact, and the comprehensive satisfaction score.

5. The method of claim 4, wherein, The attribution analysis of the abnormal factors to determine their impact on the overall satisfaction score includes: Obtain the historical mapping relationship between the abnormal factors and the overall satisfaction score; Based on the historical mapping relationship, the update priority of the abnormal factors is determined; The degree of impact is determined based on the update priority.

6. The method of claim 4, wherein, The step of determining the target update strategy based on the abnormal factors, the degree of impact, and the comprehensive satisfaction score includes: Based on the aforementioned anomalies, multiple candidate update strategies are generated; Based on the degree of impact and the comprehensive satisfaction score, simulation verification is performed on each candidate update strategy to simulate the satisfaction improvement effect corresponding to the multiple candidate update strategies. Based on the improvement in satisfaction, the target update strategy is determined from the plurality of candidate update strategies.

7. The method of claim 1, wherein, After determining the overall satisfaction score for the atomic service, the process also includes: Obtain the service call result and service call target of the atomic service; When the service call result is a service call failure and / or the overall satisfaction score is lower than a preset score threshold, the atomic service is interrupted, and the service call target is achieved based on a preset intelligent agent.

8. An atomic service update apparatus characterized by comprising: include: The data acquisition module is used to acquire user satisfaction feedback data and service operation data for atomic services; The satisfaction rating module is used to determine the overall satisfaction rating of the atomic service based on the data type of the satisfaction feedback data. The strategy update module is used to determine the target update strategy for the atomic service based on the overall satisfaction score and the service operation data. The service update module is used to update the atomic service based on the target update strategy.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the atomic service update method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. When the computer program is executed by a processor, it implements the atomic service update method as described in any one of claims 1 to 7.

Images