CN120086052A - Defect diagnosis and location method, system and storage medium - Google Patents
Defect diagnosis and location method, system and storage medium Download PDFInfo
- Publication number
- CN120086052A CN120086052A CN202510520412.6A CN202510520412A CN120086052A CN 120086052 A CN120086052 A CN 120086052A CN 202510520412 A CN202510520412 A CN 202510520412A CN 120086052 A CN120086052 A CN 120086052A
- Authority
- CN
- China
- Prior art keywords
- data
- agent
- rule
- reasoning
- characteristic information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/079—Root cause analysis, i.e. error or fault diagnosis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Prevention of errors by analysis, debugging or testing of software
- G06F11/3668—Testing of software
- G06F11/3672—Test management
- G06F11/3684—Test management for test design, e.g. generating new test cases
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Prevention of errors by analysis, debugging or testing of software
- G06F11/3668—Testing of software
- G06F11/3672—Test management
- G06F11/3688—Test management for test execution, e.g. scheduling of test suites
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The application relates to a defect diagnosis and positioning method, which is characterized in that data of each layer of a software system are collected by a data collection agent, the data of the software system are matched by a rule matching agent according to preset rules and a case library, a rule error mode is identified and determined, the data are identified and detected by an abnormality detection agent based on a machine learning algorithm, an abnormality fluctuation mode is identified and determined, finally, the defect position is determined by diagnosing and reasoning according to the identification result of a diagnosis reasoning agent combining the rule error mode and the abnormality fluctuation mode, different abnormality modes of the software system are detected through a plurality of agent cooperated drivers, the defect position is finally positioned by integrating the identification result reasoning diagnosis of the abnormality mode, and the intelligent diagnosis and accurate positioning of software testing defects are facilitated, and the efficiency and quality of software testing are improved.
Description
Technical Field
The present application relates to the field of software testing technologies, and in particular, to a defect diagnosis and positioning method, system, and storage medium.
Background
Along with the continuous expansion of the scale of the software system and the improvement of the complexity, the positioning and diagnosis of software defects become an important link for guaranteeing the stability and the reliability of the system. The existing software test defect diagnosis method mainly depends on preset rules and fixed test cases, testers usually need to manually define various possible defect modes before testing, and compare and analyze data in the running process of the system according to the modes, and particularly when processing a large-scale software system, a great deal of time and manpower are required, and flexibility and adaptability are lacked.
In the prior art, diagnosis of a software system is usually performed sequentially, and all test links are mutually independent and lack of effective cooperation and communication, so that the diagnosis process is low in efficiency. The data collected during the software testing process may have noise, missing values, or false information, which all affect the accuracy of the defect diagnosis.
In view of the above, there are many problems in the prior art for testing the defects of the software system, and a defect diagnosis and positioning method is needed to break through the bottleneck of the prior art.
Disclosure of Invention
The application provides a defect diagnosis and positioning method, a system and a storage medium, which are used for solving the problems of low efficiency in the defect diagnosis process and insufficient accuracy in positioning defect positions in the prior art.
In a first aspect, the present application provides a defect diagnosis and localization method, including:
Defining a plurality of agents for collaborative driving to test a software system, the agents including a data acquisition agent, a rule matching agent, an anomaly detection agent, and a diagnostic reasoning agent;
collecting data from each layer of the software system according to the data acquisition agent to obtain data characteristic information;
Matching the data characteristic information according to a preset rule and a case library by the rule matching agent, and identifying and determining a rule error mode;
identifying and detecting the data characteristic information based on a machine learning algorithm according to the abnormality detection intelligent agent, and identifying and determining an abnormality fluctuation mode;
And carrying out diagnosis and reasoning according to the diagnosis and reasoning agent by combining the rule error mode and the recognition result of the abnormal fluctuation mode, and determining the defect position.
Optionally, the collecting data from each layer of the software system according to the data collecting agent to obtain data characteristic information includes:
collecting data from various levels of the software system, the data including execution code, intermediate results, configuration information, user input data;
Preprocessing the data and extracting characteristic information of the data to obtain data characteristic information, wherein the data characteristic information comprises statistical characteristics, structural characteristics and semantic characteristics.
Optionally, the rule error mode includes a logic error mode and an abnormal data mode, the matching the data feature information according to the rule matching agent and a case base by a preset rule, identifying and determining the rule error mode includes:
Matching the data characteristic information according to the preset rule, and determining a logic error mode when the data characteristic information is identified to be not in accordance with the preset rule;
and matching the data characteristic information with the case library, and determining an abnormal data mode when the data characteristic information is identified to be matched with the defect case in the case library.
Optionally, the abnormal fluctuation mode includes a performance abnormal mode and an abnormal data mode, the data characteristic information is identified and detected based on a machine learning algorithm according to the abnormal detection agent, and the abnormal fluctuation mode is identified and determined, including:
based on a machine learning algorithm, learning the distribution and change rules of preset performance indexes, identifying the performance abnormality in the data characteristic information, and determining a performance abnormality mode;
And identifying abnormal data of the data characteristic information based on a machine learning algorithm, and determining an abnormal data mode, wherein the abnormal data comprises mutation and deviation from a normal range.
Optionally, the determining the defect position according to the diagnosis reasoning agent and combining the rule error mode and the recognition result of the abnormal fluctuation mode includes:
analyzing the association between the rule error mode and the recognition result of the abnormal fluctuation mode to obtain fusion information;
And carrying out diagnosis reasoning on the fusion information based on a reasoning algorithm, and determining the defect position.
Optionally, the reasoning algorithm includes rule reasoning, case reasoning and model reasoning, and the diagnosing reasoning is performed on the fusion information based on the reasoning algorithm according to the diagnosing reasoning agent, so as to determine a defect position, including:
Deducing the fusion information according to a preset rule based on the rule reasoning, and judging a first position range of the defect;
retrieving from a case library based on the case reasoning and the first position range of the defect, and judging a second position range of the defect;
inputting the model reasoning and the second position range of the defect into a preset simulation model, and positioning the specific position of the defect.
Optionally, the identifying the abnormal data of the data characteristic information based on the machine learning algorithm, and determining the abnormal data mode include:
Inputting the data characteristic information into a trained logistic regression model based on a supervised learning algorithm, classifying the data characteristic information, calculating abnormal probability and screening first abnormal data;
clustering the data characteristic information based on an unsupervised learning algorithm, analyzing data distribution, and identifying second abnormal data which is not detected by the supervised learning algorithm;
and fusing the first abnormal data and the second abnormal data to determine an abnormal data mode.
In a second aspect, the present application provides a defect diagnosis and localization system, the system comprising:
defining an agent module for defining a plurality of agents for co-driving to test a software system, the agents including a data acquisition agent, a rule matching agent, an anomaly detection agent, and a diagnostic reasoning agent;
the data acquisition module is used for collecting data from each layer of the software system according to the data acquisition agent to obtain data characteristic information;
the rule matching module is used for matching the data characteristic information according to a preset rule and a case base according to the rule matching agent, and identifying and determining a rule error mode;
The anomaly detection module is used for identifying and detecting the data characteristic information based on a machine learning algorithm according to the anomaly detection agent, and identifying and determining an anomaly fluctuation mode;
and the diagnosis reasoning module is used for carrying out diagnosis reasoning according to the diagnosis reasoning agent and combining the rule error mode and the recognition result of the abnormal fluctuation mode to determine the defect position.
In a third aspect, an embodiment of the present application provides a terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing a method as described above when executing the computer program.
In a fourth aspect, embodiments of the present application provide a computer readable storage medium storing a computer program which, when executed by a processor, implements a method as described above.
Compared with the prior art, the technical scheme provided by the embodiment of the application has the advantages that the data of each layer of the software system is collected by the data collection agent, so that the comprehensive and efficient collection of various related data is facilitated. The rule matching agent matches the data of the software system according to preset rules and a case library, identifies and determines a rule error mode, the anomaly detection agent identifies and detects the data based on a machine learning algorithm, identifies and determines an anomaly fluctuation mode, and enables different agents to be respectively responsible for testing different modules or service flows through the cooperative driving of a plurality of agents, so that different anomaly modes of the software system are detected, accurate identification of software defect characteristics and mode difficulties are facilitated, and efficiency in the defect diagnosis process is improved. And finally, carrying out diagnosis and reasoning according to the diagnosis and reasoning agent and the recognition result of the rule error mode and the abnormal fluctuation mode, and determining the defect position, thereby being beneficial to realizing intelligent diagnosis and accurate positioning of the software testing defects and improving the efficiency and quality of the software testing.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.
One or more embodiments are illustrated by way of example and not limitation in the figures of the accompanying drawings, in which like references indicate similar elements, and in which the figures of the drawings are not to be taken in a limiting sense, unless otherwise indicated.
FIG. 1 is a schematic flow chart of a defect diagnosis and localization method according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a defect diagnosis and localization system according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a computer-readable storage medium according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The following disclosure provides many different embodiments, or examples, for implementing different structures of the invention. In order to simplify the present disclosure, components and arrangements of specific examples are described below. They are, of course, merely examples and are not intended to limit the invention. Furthermore, the present invention may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed.
FIG. 1 is a schematic diagram of a defect diagnosis and localization method according to an embodiment of the present application, the method includes:
S100, defining a plurality of agents, wherein the agents are used for being driven cooperatively to test a software system, and the agents comprise a data acquisition agent, a rule matching agent, an abnormality detection agent and a diagnosis reasoning agent.
In the embodiment of the application, the Agent (Agent) refers to a program or entity capable of autonomously sensing an environment and making a decision, and is commonly found in an AI, a distributed system or a software architecture, and the Agent (INTELLIGENT AGENT) refers to a program or entity capable of autonomously learning, sensing, making a decision and acting, such as a chat robot or an autopilot algorithm. And carrying out information interaction and cooperative work among the agents through a defined communication protocol. They can request and provide data, shared knowledge, and reasoning results from each other, depending on the task requirements. For example, the data collection Agent sends the collected data to the rule matching Agent and the abnormality detection Agent for analysis, and the diagnosis reasoning Agent performs comprehensive judgment according to the rule matching and the abnormality detection results. According to the application, different intelligent agents are respectively responsible for testing different modules or service flows through the cooperative driving of the plurality of intelligent agents, so that different abnormal modes of the software system are detected, accurate identification of software defect characteristics and difficult mode is facilitated, and efficiency in the defect diagnosis process is improved.
And S200, collecting data from each layer of the software system according to the data acquisition agent to obtain data characteristic information.
Specifically, the collecting data from each layer of the software system according to the data collecting agent to obtain data characteristic information includes:
S201, collecting data from each layer of the software system, wherein the data comprises execution codes, intermediate results, configuration information and user input data.
In the embodiment of the application, the data acquisition agent is responsible for collecting related data from different components, modules and running environments of the software system, wherein the data comprise not only execution codes and intermediate results of the software, but also configuration information of the system, user input data and the like, such as system logs, user operation records, performance indexes and the like. By collecting the multi-source data, more comprehensive software running state information can be obtained.
S202, preprocessing the data and extracting feature information of the data to obtain data feature information, wherein the data feature information comprises statistical features, structural features and semantic features.
The data collected during the software testing process may have noise, missing values or error information, which all affect the accuracy of defect diagnosis, and in the embodiment of the application, the collected raw data is subjected to preprocessing operations such as cleaning, conversion, feature extraction and the like. The cleaning data can remove noise and error data, improve the quality of the data, unify the data with different formats by converting the data, facilitate the subsequent analysis, and the feature extraction is to extract valuable features for defect diagnosis from mass data, reduce the dimension and complexity of the data, and facilitate the improvement of the accuracy of defect diagnosis.
In an embodiment of the application, features related to software defects are extracted from the preprocessed data by using machine learning and data mining techniques. The features can be statistical features (such as mean values, variances, frequencies and the like) of the data, structural features (such as calling relations of codes, dependency relations of the data and the like) or semantic features (such as notes of codes, variable names and the like), and complex software data can be converted into feature vectors which are easy to analyze and understand by extracting the features, so that data processing is facilitated, and data feature information is obtained.
In the application, the rule matching agent and the abnormality detection agent can both use the extracted characteristics to perform pattern recognition, wherein the rule matching agent matches the characteristic vectors according to a preset rule base to find out the data conforming to the known defect pattern, and the abnormality detection agent recognizes abnormal characteristics different from the normal pattern through a trained model, and the abnormal characteristics possibly suggest potential defects in the software, and detect different abnormal patterns of the software system from the palace, thereby being beneficial to accurately recognizing the defect characteristics and pattern difficulties of the software and improving the efficiency in the defect diagnosis process, and being described by the following steps S300 and S400 one by one.
And S300, matching the data characteristic information according to a preset rule and a case library by the rule matching agent, and identifying and determining a rule error mode.
In the embodiment of the application, the rule matching agent is used for matching the collected data according to the preset rule and mode so as to find out potential abnormal conditions. The rule matching agent is based on preset rule and mode construction, and analysis and summary can be carried out according to past fault data and defect reports of the software system.
Specifically, the rule error mode includes a logic error mode and an abnormal data mode, the matching of the data characteristic information according to the rule matching agent and a case base by a preset rule, and the identifying and determining the rule error mode include:
S301, matching the data characteristic information according to the preset rule, and determining a logic error mode when the data characteristic information is identified to be not in accordance with the preset rule.
In the embodiment of the application, the logic error mode comprises condition judgment errors, circulation logic errors and the like in codes. For example, if the condition judgment statement wrongly writes the discount calculation logic into a section of code for calculating the total price of the commodity, the discount calculation logic is wrongly applied, and the discount calculation logic belongs to the logic error mode. Rule matching agents can identify such patterns by checking the consistency of the code logic structure with preset correct logic rules.
In the embodiment of the application, the data characteristic information is matched with the preset rule, and when the data characteristic information is not matched with the preset rule, the condition judgment error, the logic circulation error and the like of the data are indicated, the data belong to the logic error mode is determined, and meanwhile, the logic error result is detected.
S302, matching the data characteristic information with the case library, and determining an abnormal data mode when the data characteristic information is identified to be matched with a defect case in the case library.
In the embodiment of the application, the abnormal data mode comprises covering data missing, data type error, data value exceeding a reasonable range and the like. In database operation, if a field is an integer type but character string data is inserted, or a key data is lost in the transmission process, the data is in a data exception mode.
For example, in a certain e-commerce system, when a situation that the order state is not updated for a long time after a user places an order occurs, the situation that the order state cannot be changed in time due to the fact that the inventory data update delay is caused when the order processing module interacts with the inventory module is found through investigation. Based on this, rules may be set that determine an anomaly when an order is generated for more than a certain time (e.g., 10 minutes) and the inventory module does not return confirmation update information.
Industry specifications and best practices, on the other hand, are also important sources of rules. Taking financial software as an example, according to the data security specification of the financial industry, user sensitive information (such as an identity card number and a bank card number) is required to be encrypted in the transmission and storage processes. Therefore, rules can be set to check whether the sensitive information field adopts an encryption algorithm (such as an AES encryption algorithm) conforming to the industry standard in the data transmission and storage process, and if not, the sensitive information field is regarded as abnormal. After collecting the data in the running process of the software system, the rule matching agent starts working. The method comprises the steps of preprocessing data and extracting key information and characteristics. Once the rule matching agent finds that the data does not match the case data in the database, a potential anomaly is found and an anomaly data pattern is determined.
It should be noted that the logic error pattern that can be identified by the rule matching agent is mainly determined according to whether it matches with a preset rule, the abnormal data pattern is mainly determined according to whether it matches with case data in a preset database, and both abnormal patterns are determined by a preset matching mechanism.
S400, identifying and detecting the data characteristic information based on a machine learning algorithm according to the abnormality detection agent, and identifying and determining an abnormality fluctuation mode.
The abnormal fluctuation mode comprises a performance abnormal mode and an abnormal data mode, the data characteristic information is identified and detected based on a machine learning algorithm according to the abnormal detection agent, and the abnormal fluctuation mode is identified and determined, and the abnormal fluctuation mode comprises the following steps:
s401, learning distribution and change rules of preset performance indexes based on a machine learning algorithm, identifying performance anomalies in the data characteristic information, and determining a performance anomaly mode.
In the embodiment of the application, the abnormal performance mode is represented by overlong software response time, overhigh or overlow resource utilization rate and the like. When the software is processing a large amount of data, the CPU usage continues to be over 90% and the response time is prolonged from the original seconds to tens of seconds, which is a performance anomaly mode. The anomaly detection agent learns the distribution and change rules of the normal performance index by using a machine learning algorithm, thereby identifying such anomalies.
S402, identifying abnormal data of the data characteristic information based on a machine learning algorithm, and determining an abnormal data mode, wherein the abnormal data comprises mutation and deviation from a normal range.
In the embodiment of the application, the abnormal data mode comprises covering data missing, data type error, data value out of a reasonable range and the like. The anomaly detection agent can check and identify the type, the integrity and the value range of the data.
Specifically, the machine learning algorithm is mainly divided into supervised learning and unsupervised learning, and the identifying of the abnormal data of the data characteristic information based on the machine learning algorithm, and determining the abnormal data mode include:
s4021, inputting the data characteristic information into a trained logistic regression model based on a supervised learning algorithm, classifying the data characteristic information, calculating abnormal probability and screening first abnormal data.
In the embodiment of the application, the process of supervising the learning algorithm is as follows, firstly, the abnormality detection agent will perform model training by using the existing marking data, namely, the data set containing normal and abnormal data samples. For example, using a logistic regression model, the model learns the feature boundaries of normal data and abnormal data during the training process. Taking network traffic data as an example, the size of the network traffic, the request frequency, etc. are all in a certain range under normal conditions. When the rule matching agent matches according to the rule or the case library, the matching does not accord with the triggering network flow abnormal rule, the abnormality detection agent inputs the collected network flow related data, such as the request number per second, the bandwidth occupation and other characteristics, into the trained logistic regression model. If the model prediction result is the abnormal category, the abnormal condition is further confirmed, and meanwhile, the severity of the abnormality can be judged according to the probability value output by the model.
Based on a logistic regression model trained by a supervised learning algorithm, a preset error value is introduced according to the error value between the data characteristic information and the standard data characteristic information, and when the error value is larger than the preset error value, the data is marked as first abnormal data.
S4022, clustering the data characteristic information based on an unsupervised learning algorithm, analyzing data distribution, and identifying second abnormal data which is not detected by the supervised learning algorithm.
In embodiments of the present application, unsupervised learning typically employs a clustering algorithm such as DBSCAN (application of density based spatial clustering). It clusters data points by density, the data points that are connected by density are divided into the same cluster, and the data points in the low density area are regarded as abnormal points. When analyzing the software running log, the log record contains information such as a time stamp, an operation type, a module name and the like, and the abnormality detection agent clusters the log data by using a DBSCAN algorithm. If a log record cannot be divided into any existing clusters or is in a low density area at the edge of a cluster, it indicates that the log may correspond to an abnormal situation, and it may be that the software has unexpected operations or errors.
Automatic encoders are also a powerful unsupervised learning tool. It consists of an encoder that maps the input data to a low-dimensional representation and a decoder that reconstructs the low-dimensional representation back into the original data form. Under normal conditions, the reconstruction error of the normal data by the automatic encoder is small. When the rule matching agent triggers an abnormality, the abnormality detection agent inputs relevant data into the automatic encoder, and if the reconstruction error exceeds a set threshold value, the data has a larger difference from the normal data mode, and is likely to be abnormal data. For example, in image recognition software, if an excessive error in reconstruction of the image data by the automatic encoder is detected, this may mean that the image data is tampered with or that the image recognition module of the software fails.
In one possible implementation, a clustering algorithm is adopted based on an unsupervised learning algorithm, clustering is performed according to data characteristic information and standard data characteristic information, data points with connected densities are divided into the same cluster, and data points in a low-density area are regarded as second abnormal data. In another possible implementation, an automatic encoder is used to input data characteristic information based on an unsupervised learning algorithm, and if the reconstruction error exceeds a set threshold, the data is considered as second abnormal data if the reconstruction error indicates that the data has a large difference from the normal data mode.
It should be noted that, in the application, the first abnormal data can be obtained only based on the supervised learning algorithm, the second abnormal data can be obtained only based on the unsupervised learning algorithm, and the second abnormal data which is not detected by the supervised learning algorithm can be identified on the basis of screening the first abnormal data by combining the supervised learning algorithm and the unsupervised learning algorithm, so that the abnormal data can be detected more accurately, and the accuracy of defect detection can be improved.
S500, performing diagnosis and reasoning according to the diagnosis and reasoning agent and combining the rule error mode and the recognition result of the abnormal fluctuation mode, and determining the defect position.
In the embodiment of the application, the diagnosis reasoning agent is used for comprehensively matching analysis results of the rule matching agent and the abnormality detection agent, and the reasoning algorithm is used for deep diagnosis to determine possible reasons and positions of the defects. The interaction flow of the plurality of intelligent agents is summarized as that the data received by the diagnosis reasoning intelligent agents are collected by the intelligent agents from multiple layers and multiple components of the software system, and the system log, the user operation record, the performance index and the like are covered. The system log records key events such as function call and error information in the running process of the software, the user operation record reflects the interaction behavior of the user and the software, the defects related to the user operation are found, and the performance index reflects the resource use condition and response speed in the running process of the software. The rule matching agent matches the collected data according to preset rules and modes to obtain potential abnormal conditions. For example, in a section of user login verification code, the preset user name length of the rule matching agent is between 6 and 20 bits, and if the input user name length is detected to be not in accordance with the rule, the abnormal information is transmitted to the diagnosis reasoning agent. The abnormality detection agent uses a machine learning or statistical analysis method to identify abnormal patterns of data, such as data mutation, deviation from normal range, etc. When the use condition of the software memory is monitored, the abnormality detection agent discovers that the memory occupation rapidly rises in a short time and exceeds the normal fluctuation range, and then the abnormality mode information is sent to the diagnosis reasoning agent.
Specifically, the diagnosing and reasoning agent performs diagnosing and reasoning according to the identifying result of the rule error mode and the abnormal fluctuation mode, and determining the defect position includes:
s501, analyzing the association between the rule error mode and the recognition result of the abnormal fluctuation mode to obtain fusion information.
In the embodiment of the application, after each agent completes the analysis task of the agent, the result is sent to the diagnosis reasoning agent. The diagnosis reasoning agent fuses the information, and the views and evidences of different agents are comprehensively considered. For example, rule matching agents find that certain code segments may have logical errors, anomaly detection agents detect anomalous fluctuations in system performance, and diagnostic reasoning agents can combine the two information to further analyze the correlation between them.
S502, carrying out diagnosis and reasoning on the fusion information based on a reasoning algorithm, and determining the defect position.
In the embodiment of the application, the diagnosis reasoning agent uses various reasoning algorithms, such as rule-based reasoning, case-based reasoning, model-based reasoning and the like, to conduct deep analysis and reasoning on the fused information. By means of the reasoning algorithms, the possible range of the defect can be gradually narrowed, and the specific position and the cause of the defect can be determined. For example, rule-based reasoning can derive new conclusions from known information according to preset rules and logic relationships, and case-based reasoning can quickly locate current problems by referring to previous experience of processing similar defects.
The diagnosis and reasoning agent uses various reasoning algorithms, such as rule-based reasoning, case reasoning, model reasoning and the like, to conduct deep analysis and reasoning on the fused information. By means of the reasoning algorithms, the possible range of the defect can be gradually narrowed, and the specific position and the cause of the defect can be determined. For example, rule-based reasoning can derive new conclusions from known information according to preset rules and logic relationships, and case-based reasoning can quickly locate current problems by referencing past experience of processing similar defects.
Specifically, the diagnosing and reasoning agent performs diagnosing and reasoning on the fusion information based on a reasoning algorithm according to the diagnosing and reasoning agent, and determining the defect position includes:
S5021, deducing the fusion information according to a preset rule based on the rule reasoning, and judging a first position range of the defect.
In the embodiment of the application, the rule reasoning working principle is that the rule-based reasoning carries out logic deduction according to a series of preset rules. These rules are typically expressed in terms of "if (condition), then (conclusion)" and are a summary of knowledge of the software domain and common defect patterns. In the diagnosis process, the diagnosis reasoning agent matches the analysis results provided by other agents with rules in the rule base. If the condition of a certain rule is met, a corresponding conclusion can be deduced, and a first position range of the defect, namely a position range where the defect possibly appears, is determined, wherein the first position range is larger than the second position range and the specific position of the defect. For example, when detecting the file read/write function of software, the rule is set to "if the file read operation returns an error code and the file path exists and the authority is correct, then it may be a file format incompatibility problem". When the rule matching agent finds that the file reading operation is wrong, and the information provided by the data acquisition agent indicates that the file path and the authority are free of problems, the diagnosis reasoning agent can preliminarily judge that the problems are likely to occur in the aspect of the file format according to the rule.
In the embodiment of the application, for the logic error mode, after the diagnosis reasoning agent receives the logic error mode information identified by the rule matching agent, a rule-based reasoning method is applied. And further analyzing possible reasons for error generation according to preset logic rules and business logic of the program. For example, if the rule matching agent finds that the condition judgment is wrong in a certain section of order processing code, the diagnosis reasoning agent deduces that the order processing flow is abnormal due to the setting error of the boundary value of the condition judgment according to the normal business logic rule of order processing, such as the change condition of the order state, the inventory deduction logic of goods, and the like, so as to further determine the first position range of the defect.
S5022, searching from a case library based on the case reasoning and the first position range of the defect, and judging the second position range of the defect.
In the embodiment of the application, the working principle of case-based reasoning is that the current problems are compared with similar cases which are solved in the past, and the solutions of the similar cases are used for processing the current problems. The diagnostic reasoning agent will first retrieve from the case base the historical cases that are similar to the current defect features. Each case in the case library contains information such as problem description, solution and final effect. The method comprises the steps of finding the most similar cases or case sets by calculating the similarity of the current problem and the cases in the case library, then adjusting and applying according to the solutions of the similar cases and combining the specific situations of the current problem, and determining the position range where the defect occurs with high probability, wherein the second position range is smaller than the first position range and comprises the specific position of the defect.
For example, when diagnosing a flicker problem in a graphics rendering software, the diagnostic reasoning agent searches a case library for similar flicker cases in which the flicker has been processed in the past, the cases being caused by too low a driver version of the graphics card. The diagnosis reasoning intelligent body refers to the case, firstly checks the display card driving condition of the current system, and decides whether the display card driving needs to be updated according to the actual condition so as to solve the current problem.
The method is applied to the embodiment of the application, and a model-based reasoning method is adopted for diagnosing and reasoning intelligent bodies aiming at abnormal performance modes. And constructing a software performance model, and simulating the performance of the software under different load conditions. After the abnormality detection agent identifies the performance abnormality mode, the diagnosis reasoning agent inputs the actual performance data into the model, and compares the difference between the model output and the actual situation. If the model shows that the response time does not exceed a certain threshold under the current load, but the actual response time is very long, determining whether the database index is unreasonably set to cause the overlong query time or the efficiency of a certain algorithm in processing a large amount of data is low by analyzing the influence of each component in the model on the performance, such as database query efficiency, algorithm complexity and the like, so as to locate the second position range of the defect.
S5023, inputting the model reasoning and the second position range of the defect into a preset simulation model, and positioning the specific position of the defect.
In the embodiment of the application, the working principle of model reasoning is that the model-based reasoning simulates the normal behavior and the running process of a system by constructing a model of a software system. The model may be a structural model, a functional model, a behavioral model, or the like, reflecting relationships among components of the software system, functional implementation of the system, and behavioral changes under different conditions. The diagnostic reasoning agent inputs the collected software operation data into the model to perform simulation operation, and then compares the output result of the model with the actual observed software operation condition. If there are differences, it is indicated that the software system may have defects, and the specific location and cause of the defects are determined by analyzing the differences.
For example, in the test of a network communication software, a network communication model is constructed, which includes links such as sending, transmitting, receiving and error processing of data packets. When the problem of too high data transmission delay occurs in actual network communication, the relevant network parameters and the data transmission condition input model are simulated. If the model simulation result shows that the high delay should not occur under normal conditions, links such as a data packet transmission path, a congestion control mechanism and the like in the model are further analyzed, and the fact that congestion occurs in a certain network node or a transmission link is possibly failed is judged by combining the actual network topology and the equipment state.
Application in embodiments of the present application, for data anomaly patterns, a combination of rule-based and case-based reasoning approaches. Rule-based reasoning is used to check whether a data exception violates a data rule of a database or software system, and to determine the cause of a data type error or an out-of-range data value. Case-based reasoning is to search cases which are used for processing similar data anomalies from a case library and reference the solution. If the abnormality detection agent finds the problem of data missing, the diagnosis reasoning agent first judges the possible influence of the data missing on the service function according to the data integrity rule, and then refers to the processing method of similar data missing cases in the case library, such as recovering from backup data, checking a data transmission link and the like, and determines the specific position and the repairing scheme of the defect.
By way of example of defect diagnosis and localization, the above-mentioned processes S5021-5023 are processes of gradually narrowing the defect position range, (1) rule-based reasoning narrowing, in which the diagnostic reasoning agent performs reasoning according to a preset rule after receiving information of other agents. In the test of the file reading function of the software, if the rule matching agent finds that the file reading operation returns an error code and the file path and authority information provided by the data collecting agent are displayed normally, according to the rule of 'the file reading error and the path authority is normal, which may be the incompatibility of the file format', the diagnosis reasoning agent can judge that the defect may appear in the code module related to the file format processing at first step, and the investigation range of the whole file reading function originally is reduced to a file format processing part.
(2) Case-based reasoning is further focused on the fact that the diagnostic reasoning agent can be more accurately narrowed down by means of previous experience of processing similar defects. When the problem of the interface jam of the software is processed, if the rule-based reasoning preliminary determination is related to the graphic rendering module, the diagnosis reasoning agent searches the past similar interface jam cases from the case library, and the finding is mostly caused by improper texture loading mode. With reference to the cases, the diagnosis reasoning agent further focuses the investigation key on the code part of the texture loading in the graphic rendering module, and performs deep examination on parameter setting, loading sequence and the like of the texture loading function.
(3) The reasoning accurate positioning based on the model is that the diagnosis reasoning intelligent body can accurately position the defect by constructing a software system model and simulating the normal running condition of the system. In a network communication software, a model is built that includes data packet transmission, reception, and processing flows. When the problem of too high data transmission delay occurs, the diagnosis reasoning agent inputs the actual network parameters and the transmission data into the model. If the model simulation result shows that the high delay does not occur in the current network environment, links such as a data packet transmission path, a network congestion control mechanism and the like in the model are analyzed, and the actual network topology and the equipment state are combined, so that the routing configuration error of a certain network node is accurately positioned, the data packet transmission is caused to have roundabout, and the delay is too high.
(4) The method for determining the specific position of the defect comprises the steps of diagnosing and reasoning an agent to generate a defect report according to a reasoning result, describing the defect characteristics, possible reasons and influence ranges in detail, and providing repair guidance for developers. The code file, function name, specific code line where the defect is located are explicitly indicated in the defect report. In an order processing module of an e-commerce system, a diagnosis reasoning agent determines that the total price calculation error of an order is caused by a multiplication error in an order amount calculation function, and a code file path, a function name and a specific code line number with errors of the function are recorded in a defect report. The diagnostic reasoning agent can determine the module or system component where the defect is located by combining the structural information of the software system, such as module division, inheritance relation of classes, function call hierarchy and the like. In a large enterprise-level application, if a certain business logic error is found, the error is determined to belong to a cost calculation sub-assembly under a financial accounting module through analyzing a system architecture, so that a developer can conveniently and rapidly locate and repair the defect.
The method also comprises the step of feeding back the defect positioning result to a developer for repairing. Meanwhile, the system can optimize and adjust rules, models and algorithms of the multi-agent system according to the repairing result and actual conditions, and improves the diagnosis accuracy and efficiency of the system, so that defects can be found and positioned better in subsequent software tests.
Compared with the prior art, the method has the advantages that (1) the traditional method is difficult to collect various relevant data comprehensively and efficiently in the face of a complex software system, and the method is solved by collecting data from multiple layers and multiple components through a data collecting agent. Such as the problem of specially detecting boundary values, and some focus on logical error checking, thereby improving the defect discovery rate. (2) The software defect characteristics and the mode are difficult to accurately identify, the characteristics are extracted by utilizing technologies such as machine learning and the like, and the mode identification is carried out by rule matching and anomaly detection intelligent bodies. The method for cooperatively driving the multiple agents can enable different agents to be respectively responsible for testing different modules or service flows, and improves testing efficiency and accuracy through cooperative work. (3) The multi-agent cooperation can realize the monitoring and defect diagnosis of different nodes and services in a distributed environment, discover and locate possible faults and defects in time and feed back to a user for optimization.
As shown in fig. 2, fig. 2 is a schematic structural diagram of a defect diagnosis and positioning system according to an embodiment of the present application, including:
A define agents module 610 for defining a plurality of agents for co-driving to test the software system, the agents including a data acquisition agent, a rule matching agent, an anomaly detection agent, and a diagnostic reasoning agent;
A data acquisition module 620, configured to collect data from each layer of the software system according to the data acquisition agent, and obtain data feature information;
the rule matching module 630 is configured to match the data feature information according to a preset rule and a case library by using the rule matching agent, and identify and determine a rule error pattern;
The anomaly detection module 640 is configured to identify and detect the data feature information based on a machine learning algorithm according to the anomaly detection agent, and identify and determine an anomaly fluctuation mode;
and the diagnosis reasoning module 650 is used for performing diagnosis reasoning according to the diagnosis reasoning agent and combining the rule error mode and the recognition result of the abnormal fluctuation mode to determine the defect position.
As shown in fig. 3, fig. 3 is a schematic structural diagram of a computer readable storage medium according to an embodiment of the present application. The computer-readable storage medium 700 of this embodiment includes a server 710 (only one is shown in fig. 3), a client 720, and a data recovery program 721 stored in the client 720 and executable on the at least one client 720, the client 720 executing the data recovery program 721 to send a request to the server 710, and the server 710 feeding back the result to implement the steps in the above-described method embodiments.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units and modules are only for distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other manners. For example, the apparatus/terminal device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical function division, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth. It should be noted that the computer readable medium contains content that can be appropriately scaled according to the requirements of jurisdictions in which such content is subject to legislation and patent practice, such as in certain jurisdictions in which such content is subject to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.
The present application may also be implemented by a computer program product for implementing all or part of the steps of the above embodiments of the method, when the computer program product is run on a terminal device, for enabling the terminal device to execute the steps of the above embodiments of the method.
The above embodiments are only for illustrating the technical solution of the present application, and are not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications and substitutions can be made to the technical solutions described in the foregoing embodiments or equivalent substitutions can be made to some technical features thereof, and these modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. A method for defect diagnosis and localization, the method comprising:
Defining a plurality of agents for collaborative driving to test a software system, the agents including a data acquisition agent, a rule matching agent, an anomaly detection agent, and a diagnostic reasoning agent;
collecting data from each layer of the software system according to the data acquisition agent to obtain data characteristic information;
Matching the data characteristic information according to a preset rule and a case library by the rule matching agent, and identifying and determining a rule error mode;
identifying and detecting the data characteristic information based on a machine learning algorithm according to the abnormality detection intelligent agent, and identifying and determining an abnormality fluctuation mode;
And carrying out diagnosis and reasoning according to the diagnosis and reasoning agent by combining the rule error mode and the recognition result of the abnormal fluctuation mode, and determining the defect position.
2. The method of claim 1, wherein the collecting data from the various levels of the software system according to the data collection agent to obtain data characteristic information comprises:
collecting data from various levels of the software system, the data including execution code, intermediate results, configuration information, user input data;
Preprocessing the data and extracting characteristic information of the data to obtain data characteristic information, wherein the data characteristic information comprises statistical characteristics, structural characteristics and semantic characteristics.
3. The method of claim 1, wherein the rule error patterns include a logical error pattern and an abnormal data pattern, wherein the matching agent matches the data characteristic information according to a preset rule and a case library according to the rule, and identifying and determining the rule error pattern includes:
Matching the data characteristic information according to the preset rule, and determining a logic error mode when the data characteristic information is identified to be not in accordance with the preset rule;
and matching the data characteristic information with the case library, and determining an abnormal data mode when the data characteristic information is identified to be matched with the defect case in the case library.
4. The method of claim 1, wherein the abnormal fluctuation pattern includes a performance abnormal pattern and an abnormal data pattern, wherein the identifying, detecting, identifying and determining the abnormal fluctuation pattern based on the machine learning algorithm based on the data characteristic information by the abnormality detection agent includes:
based on a machine learning algorithm, learning the distribution and change rules of preset performance indexes, identifying the performance abnormality in the data characteristic information, and determining a performance abnormality mode;
And identifying abnormal data of the data characteristic information based on a machine learning algorithm, and determining an abnormal data mode, wherein the abnormal data comprises mutation and deviation from a normal range.
5. The method of claim 1, wherein said determining a defect location based on said diagnostic reasoning agent in combination with said rule error pattern and said recognition of said abnormal surge pattern comprises:
analyzing the association between the rule error mode and the recognition result of the abnormal fluctuation mode to obtain fusion information;
And carrying out diagnosis reasoning on the fusion information based on a reasoning algorithm, and determining the defect position.
6. The method of claim 5, wherein the inference algorithm includes rule inference, case inference, and model inference, and wherein said determining the defect location based on the diagnostic inference of the fusion information by the diagnostic inference agent based on the inference algorithm includes:
Deducing the fusion information according to a preset rule based on the rule reasoning, and judging a first position range of the defect;
retrieving from a case library based on the case reasoning and the first position range of the defect, and judging a second position range of the defect;
inputting the model reasoning and the second position range of the defect into a preset simulation model, and positioning the specific position of the defect.
7. The method of claim 4, wherein the identifying the abnormal data of the data characteristic information based on a machine learning algorithm, determining an abnormal data pattern, comprises:
Inputting the data characteristic information into a trained logistic regression model based on a supervised learning algorithm, classifying the data characteristic information, calculating abnormal probability and screening first abnormal data;
clustering the data characteristic information based on an unsupervised learning algorithm, analyzing data distribution, and identifying second abnormal data which is not detected by the supervised learning algorithm;
and fusing the first abnormal data and the second abnormal data to determine an abnormal data mode.
8. A defect diagnosis and localization system, the system comprising:
defining an agent module for defining a plurality of agents for co-driving to test a software system, the agents including a data acquisition agent, a rule matching agent, an anomaly detection agent, and a diagnostic reasoning agent;
the data acquisition module is used for collecting data from each layer of the software system according to the data acquisition agent to obtain data characteristic information;
the rule matching module is used for matching the data characteristic information according to a preset rule and a case base according to the rule matching agent, and identifying and determining a rule error mode;
The anomaly detection module is used for identifying and detecting the data characteristic information based on a machine learning algorithm according to the anomaly detection agent, and identifying and determining an anomaly fluctuation mode;
and the diagnosis reasoning module is used for carrying out diagnosis reasoning according to the diagnosis reasoning agent and combining the rule error mode and the recognition result of the abnormal fluctuation mode to determine the defect position.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium storing a computer program, which when executed by a processor performs the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510520412.6A CN120086052B (en) | 2025-04-24 | 2025-04-24 | Defect diagnosis and localization method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510520412.6A CN120086052B (en) | 2025-04-24 | 2025-04-24 | Defect diagnosis and localization method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120086052A true CN120086052A (en) | 2025-06-03 |
| CN120086052B CN120086052B (en) | 2025-08-01 |
Family
ID=95845249
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510520412.6A Active CN120086052B (en) | 2025-04-24 | 2025-04-24 | Defect diagnosis and localization method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120086052B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114168580A (en) * | 2021-12-07 | 2022-03-11 | 浙江中烟工业有限责任公司 | A method to improve the efficiency of AI intelligent analysis in data anomaly detection in tobacco warehousing business |
| WO2022177182A1 (en) * | 2021-02-22 | 2022-08-25 | 주식회사 미래엔지니어링 | Artificial intelligence valve failure prediction and diagnosis system of automatic valve |
| CN117787953A (en) * | 2023-12-26 | 2024-03-29 | 合肥科大智能机器人技术有限公司 | Fault diagnosis method, equipment and storage medium for intelligent operation and maintenance of power supply |
| CN118860848A (en) * | 2024-05-13 | 2024-10-29 | 武汉纺织大学 | A method for automated closed-loop testing of real-time embedded software |
| CN119397326A (en) * | 2024-09-27 | 2025-02-07 | 天津大学 | An internal combustion engine fault diagnosis and defect detection intelligent agent system and construction method |
| CN119676054A (en) * | 2024-11-18 | 2025-03-21 | 广西东信易通科技有限公司 | A method, system, device and storage medium for diagnosing abnormal cloud communication |
| CN119691576A (en) * | 2024-12-05 | 2025-03-25 | 广州汇通国信科技有限公司 | Self-adaptive operation and maintenance root cause positioning method and system based on deep learning |
| CN119739071A (en) * | 2024-12-18 | 2025-04-01 | 中电福富信息科技有限公司 | Energy management multi-agent collaborative optimization method and system based on large model agent architecture |
-
2025
- 2025-04-24 CN CN202510520412.6A patent/CN120086052B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022177182A1 (en) * | 2021-02-22 | 2022-08-25 | 주식회사 미래엔지니어링 | Artificial intelligence valve failure prediction and diagnosis system of automatic valve |
| CN114168580A (en) * | 2021-12-07 | 2022-03-11 | 浙江中烟工业有限责任公司 | A method to improve the efficiency of AI intelligent analysis in data anomaly detection in tobacco warehousing business |
| CN117787953A (en) * | 2023-12-26 | 2024-03-29 | 合肥科大智能机器人技术有限公司 | Fault diagnosis method, equipment and storage medium for intelligent operation and maintenance of power supply |
| CN118860848A (en) * | 2024-05-13 | 2024-10-29 | 武汉纺织大学 | A method for automated closed-loop testing of real-time embedded software |
| CN119397326A (en) * | 2024-09-27 | 2025-02-07 | 天津大学 | An internal combustion engine fault diagnosis and defect detection intelligent agent system and construction method |
| CN119676054A (en) * | 2024-11-18 | 2025-03-21 | 广西东信易通科技有限公司 | A method, system, device and storage medium for diagnosing abnormal cloud communication |
| CN119691576A (en) * | 2024-12-05 | 2025-03-25 | 广州汇通国信科技有限公司 | Self-adaptive operation and maintenance root cause positioning method and system based on deep learning |
| CN119739071A (en) * | 2024-12-18 | 2025-04-01 | 中电福富信息科技有限公司 | Energy management multi-agent collaborative optimization method and system based on large model agent architecture |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120086052B (en) | 2025-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111552609B (en) | An abnormal state detection method, system, storage medium, program and server | |
| Li et al. | Deeplv: Suggesting log levels using ordinal based neural networks | |
| CN113282461B (en) | Alarm recognition method and device for transmission network | |
| CN115309575B (en) | Microservice fault diagnosis method, device and equipment based on graph convolutional neural network | |
| CN118569655B (en) | Staged data life cycle safety assessment method and system | |
| US20250016185A1 (en) | Apparatus and method for automatically analyzing malicious event log | |
| CN106776208B (en) | Fault Locating Method when a kind of running software | |
| CN113590451B (en) | Root cause positioning method, operation and maintenance server and storage medium | |
| CN113687972A (en) | Method, device and equipment for processing abnormal data of business system and storage medium | |
| CN118133962A (en) | Correlation analysis method, device and system of fault event and storage medium | |
| Ardimento et al. | Transfer Learning for Just-in-Time Design Smells Prediction using Temporal Convolutional Networks. | |
| CN119383169B (en) | IP recovery method, system, equipment and medium for regional service interruption | |
| Shi et al. | Serverrca: Root cause analysis for server failure using operating system logs | |
| CN119248617A (en) | Elasticsearch cluster fault intelligent detection system | |
| CN120086052B (en) | Defect diagnosis and localization method, system and storage medium | |
| CN119088671A (en) | Code failure warning method, device, terminal equipment and storage medium | |
| CN119025415A (en) | Detection method and device based on regression testing of application program | |
| CN119201682A (en) | Regression testing method, system, computer equipment and storage medium | |
| CN119150097A (en) | Alarm classification method and device based on large model, medium and equipment | |
| Wu et al. | Traffic sensor health monitoring using spatiotemporal graphical modeling | |
| Afshinpour et al. | Telemetry-based software failure prediction by concept-space model creation | |
| Liu et al. | LMGD: Log-Metric Combined Microservice Anomaly Detection Through Graph-Based Deep Learning | |
| RU2786934C1 (en) | Method for predicting the failure of sensor and wireless network equipment based on ontology using machine learning | |
| Al Amin | Supervised Learning for Detecting Cognitive Security Anomalies in Real-Time Log Data | |
| JP6547341B2 (en) | INFORMATION PROCESSING APPARATUS, METHOD, AND PROGRAM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |