CN115994098A - Analysis method and system for software operation abnormality based on artificial intelligence - Google Patents
Analysis method and system for software operation abnormality based on artificial intelligence Download PDFInfo
- Publication number
- CN115994098A CN115994098A CN202310282473.4A CN202310282473A CN115994098A CN 115994098 A CN115994098 A CN 115994098A CN 202310282473 A CN202310282473 A CN 202310282473A CN 115994098 A CN115994098 A CN 115994098A
- Authority
- CN
- China
- Prior art keywords
- software
- analyzed
- running
- degree
- interaction
- 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
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 26
- 238000004458 analytical method Methods 0.000 title claims abstract description 25
- 230000005856 abnormality Effects 0.000 title claims description 34
- 230000003993 interaction Effects 0.000 claims description 114
- 230000000694 effects Effects 0.000 claims description 88
- 230000002159 abnormal effect Effects 0.000 claims description 40
- 238000000034 method Methods 0.000 claims description 23
- 238000012360 testing method Methods 0.000 claims description 19
- 238000003062 neural network model Methods 0.000 claims description 13
- 238000012163 sequencing technique Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 description 23
- 238000012549 training Methods 0.000 description 18
- 238000012545 processing Methods 0.000 description 12
- 230000002155 anti-virotic effect Effects 0.000 description 9
- 238000013528 artificial neural network Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000000306 recurrent effect Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 238000011478 gradient descent method Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The invention provides an analysis method and system for software operation abnormity based on artificial intelligence, and relates to the technical field of software abnormity analysis.
Description
Technical Field
The invention relates to the technical field of software exception analysis, in particular to an artificial intelligence-based software operation exception analysis method and system.
Background
With the development of the internet, the performance of the user computer is better and the installed software is more and more. With the occurrence of functional conflicts between multiple software running simultaneously, an abnormal running condition of some software is caused, and the abnormal condition includes crashes, functional failures, clamping and the like. For example, functional conflicts sometimes occur between multiple antivirus software, resulting in a user having a computer stuck, and other software components having a functional failure, and for example, functional conflicts sometimes occur between an input method and a computer game, resulting in a user being unable to type in the game. For another example, the new version of software and the old version of software may be incompatible, resulting in the new version of software being partially dysfunctional. When abnormality occurs in software, it is necessary to perform investigation of operation abnormality caused by incompatibility between software. Most of incompatible abnormality checks among existing software are to analyze logs generated by the software in detail through a related analysis tool and identify the cause of the abnormality, but the method requires users to have enough knowledge of the log format and content of the software and to be skilled in using the related analysis tool.
Therefore, how to quickly and accurately determine incompatible software is a current urgent problem to be solved.
Disclosure of Invention
The invention mainly solves the technical problem of how to quickly and accurately determine incompatible software.
According to a first aspect, the present invention provides a method for analyzing abnormal software operation based on artificial intelligence, comprising: receiving an instruction of abnormal operation of software to be analyzed, which is sent by a user; acquiring interaction data of the software to be analyzed and a plurality of running software; determining the interaction degree of the plurality of running software and the software to be analyzed by using an interaction degree determining model based on the interaction data of the software to be analyzed and the plurality of running software; acquiring a screen recorded video before the operation abnormality of the software to be analyzed and the operation environment information of the software to be analyzed; determining the activity degrees of the plurality of running software by using an activity degree determining model based on the screen recorded video before the running abnormality of the software to be analyzed and the running environment information of the software to be analyzed; determining one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed and the activity degree of the plurality of running software; and sequentially testing the one or more associated software to determine incompatible software corresponding to the software to be analyzed.
Furthermore, the interaction degree determining model is a deep neural network model, the input of the interaction degree determining model is interaction data of the software to be analyzed and a plurality of running software, and the output of the interaction degree determining model is interaction degree of the plurality of running software and the software to be analyzed.
Furthermore, the activity level determining model is a long-short period neural network model, the input of the activity level determining model is a screen recorded video before the running abnormality of the software to be analyzed and running environment information of the software to be analyzed, and the output of the activity level determining model is the activity level of the plurality of running software.
Further, determining one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed and the activity degree of the plurality of running software; comprising the following steps: sequencing the plurality of running software according to the sequence of the interaction degree from large to small, and selecting a plurality of running software with the top N of the interaction degree ranking; sequencing the plurality of running software according to the sequence of the activity degrees from large to small, and selecting a plurality of running software with the activity degrees ranked N at the top; and determining the same software in the running software with the interaction degree of N ranked in front and the running software with the activity degree of N ranked in front as one or more associated software, wherein N is a positive integer greater than or equal to 1.
Further, the sequentially testing the one or more associated software to determine incompatible software corresponding to the software to be analyzed includes: and carrying out unloading operation on the one or more associated software in sequence, detecting the software to be analyzed after each unloading operation, and taking the associated software corresponding to the unloading operation as incompatible software if the fact that the software to be analyzed is not abnormal in operation is detected.
Still further, the method further comprises: and if the incompatible software is not found after the sequential test, notifying a user to adopt other abnormal analysis means to analyze the software to be analyzed.
According to a second aspect, the present invention provides an artificial intelligence based analysis system for software operation anomalies, comprising: an artificial intelligence based analysis system for software operational anomalies, comprising: the receiving module is used for receiving an instruction of abnormal operation of the software to be analyzed, which is sent by a user; the first acquisition module is used for acquiring interaction data of the software to be analyzed and a plurality of running software; the interaction degree determining module is used for determining interaction degrees of the plurality of running software and the software to be analyzed by using an interaction degree determining model based on interaction data of the software to be analyzed and the plurality of running software; the second acquisition module is used for acquiring a screen recorded video before the operation abnormality of the software to be analyzed and the operation environment information of the software to be analyzed; the activity degree determining module is used for determining the activity degrees of the plurality of running software by using an activity degree determining model based on the screen recorded video before the running abnormality of the software to be analyzed and the running environment information of the software to be analyzed; the associated software determining module is used for determining one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed and the activity degree of the plurality of running software; and the incompatible software determining module is used for sequentially testing the one or more associated software and determining incompatible software corresponding to the software to be analyzed.
Still further, the association software determination module is further configured to: sequencing the plurality of running software according to the sequence of the interaction degree from large to small, and selecting a plurality of running software with the top N of the interaction degree ranking; sequencing the plurality of running software according to the sequence of the activity degrees from large to small, and selecting a plurality of running software with the activity degrees ranked N at the top; and determining the same software in the running software with the interaction degree of N ranked in front and the running software with the activity degree of N ranked in front as one or more associated software, wherein N is a positive integer greater than or equal to 1.
According to a third aspect, the present invention provides an electronic device comprising: a memory; a processor; a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method described above.
According to a fourth aspect, the present invention provides a computer readable storage medium having stored thereon a program executable by a processor to implement a method as in any of the above aspects.
According to the analysis method, the system, the equipment and the medium for the software operation abnormality based on the artificial intelligence, the interaction degree determining model is used for determining the interaction degree of a plurality of operation software and a plurality of operation software based on the interaction data of the software to be analyzed and the operation software, the activity degree determining model is used for determining the activity degree of the plurality of operation software based on the screen recorded video before the operation abnormality of the software to be analyzed and the operation environment information of the software to be analyzed, one or more associated software is determined based on the interaction degree of the plurality of operation software and the operation software to be analyzed, and finally the one or more associated software is tested in sequence to determine incompatible software corresponding to the software to be analyzed, so that the incompatible software can be determined quickly and accurately.
Drawings
FIG. 1 is a schematic flow chart of an analysis method of software operation abnormality based on artificial intelligence according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of sequencing a plurality of running software based on interaction level and activity level according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an analysis system for software operation anomalies based on artificial intelligence according to an embodiment of the present invention;
fig. 4 is a schematic diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The invention will be described in further detail below with reference to the drawings by means of specific embodiments. In the following embodiments, numerous specific details are set forth in order to provide a better understanding of the present invention.
In the embodiment of the invention, an analysis method of software operation abnormality based on artificial intelligence is provided as shown in fig. 1, and the analysis method of software operation abnormality based on artificial intelligence comprises steps S1-S7:
step S1, receiving an instruction of abnormal operation of software to be analyzed, which is sent by a user.
The server can receive an instruction of abnormal operation of the software to be analyzed, wherein the abnormal operation condition comprises software crash, deadlock, partial functional failure, blocking and the like. Deadlock refers to the situation where in two or more concurrent processes of software, each process occupies a certain resource but cannot advance forward.
The instruction of the software to be analyzed, which is sent by the user, comprises a request, a time stamp and the like of the software to be analyzed, which runs abnormally. In some embodiments, the instruction of the abnormal operation of the software to be analyzed further includes memory occupation data, network connection data and the like of the software to be analyzed.
The software to be analyzed can be game software, office software, chat software, input method software and the like.
And the user can send an instruction of abnormal software operation to be analyzed to the server through a preset interface so that the server can analyze the abnormal software operation.
And S2, acquiring interaction data of the software to be analyzed and a plurality of running software.
The plurality of running software means a plurality of software in the device that is running in addition to the software to be analyzed. The plurality of running software may include incompatible software of the software to be analyzed, so that the plurality of running software needs to be analyzed to determine the incompatible software. In some embodiments, the interaction data between the software to be analyzed and the plurality of running software represents data exchanged after the plurality of running software and the software to be analyzed respectively establish a data communication path. The interaction data of the plurality of running software and the software to be analyzed respectively comprise the format, the content, the size, the interaction frequency, the coding mode, the interaction mode, the network transmission protocol, the interface standard and the like of the interaction data. In some embodiments, the interaction means may include IPC communication, socket communication, etc. In some embodiments, the interaction data of the software to be analyzed with the plurality of running software may include video data, text data, applications, and the like.
And step S3, determining the interaction degree of the plurality of running software and the software to be analyzed by using an interaction degree determining model based on the interaction data of the software to be analyzed and the plurality of running software.
The interaction degree of the plurality of running software and the software to be analyzed represents the data communication interaction degree of each of the plurality of software and the software to be analyzed when the plurality of running software and the software to be analyzed interact pairwise. In some embodiments, the interaction degree of the running software and the software to be analyzed can be used for evaluating the data communication interaction condition of the running software and the software to be analyzed. For example, the more data the running software exchanges with the software to be analyzed, the greater the degree of interaction the running software has with the software to be analyzed. For another example, the higher the frequency of data exchange of the running software with the software to be analyzed, the greater the degree of interaction of the running software with the software to be analyzed. For another example, the higher the proportion of data in the software to be analyzed that needs to be transferred through the running software, the greater the interaction degree between the running software and the software to be analyzed. The interaction degree can be a value between 0 and 1, and the greater the value is, the greater the interaction degree between the running software and the software to be analyzed is. For example, the interaction degree is 0.2, which indicates that the interaction degree of the running software and the software to be analyzed is low, and for example, the interaction degree is 0.8, which indicates that the interaction degree of the running software and the software to be analyzed is high. As an example, the interaction degree of the three running software and the software to be analyzed may be 0.2, 0.3 and 0.8 respectively.
The likelihood of incompatibility increases when there is too much information interaction between the two applications, i.e., the greater the degree of interaction. This is because the process of information interaction involves a number of aspects, including data formats, data transfer protocols, interface standards, etc., which can lead to incompatibility problems if there are inconsistencies or conflicts in the two applications. For example, if two application software needs to share data, but a part of the shared data uses different data formats or coding modes, a data parsing error or a data loss may occur, so that software functions are disabled or abnormal. For another example, if data needs to be transmitted between two application software through a network, but a portion of the transmitted data uses a different transmission protocol or interface standard, this may cause a problem that the software cannot establish a connection or communication, resulting in incompatibility.
The interaction degree determining model is a deep neural network model. The deep neural network model includes a deep neural network (Deep Neural Networks, DNN). The deep neural network model is one implementation of artificial intelligence. The deep neural network may include a plurality of processing layers, each processing layer being composed of a plurality of neurons, each neuron matrixing data. The parameters used by the matrix may be obtained by training. The deep neural network may include a recurrent neural network (Recurrent Neural Network, RNN), a convolutional neural network (Convolutional Neural Networks, CNN), a generating countermeasure network (Generative Adversarial Networks, GAN), and so on. The input of the interaction degree determining model is interaction data of the software to be analyzed and a plurality of running software, and the output of the interaction degree determining model is interaction degree of the plurality of running software and the software to be analyzed.
The interaction degree determination model can be obtained through training of training samples. The input of the training sample is the interaction data of the sample software to be analyzed and a plurality of running software, and the output label of the training sample is the interaction degree of the plurality of running software of the sample and the software to be analyzed. In some embodiments, the interaction degree determination model may be trained by a gradient descent method to obtain a trained interaction degree determination model. Specifically, according to the training sample, constructing a loss function of the interaction degree determination model, adjusting the interaction degree to determine parameters of the model through the loss function of the interaction degree determination model until the loss function value converges or is smaller than a preset threshold value, and finishing training. The loss function may include, but is not limited to, a log (log) loss function, a square loss function, an exponential loss function, a range loss function, an absolute value loss function, and the like.
And S4, acquiring a screen recorded video before the operation abnormality of the software to be analyzed and the operation environment information of the software to be analyzed.
And the screen recorded video before the abnormal operation of the software to be analyzed represents a section of video before the abnormal operation of the software to be analyzed, which is obtained by recording the screen interface. The screen recording video can be obtained by recording the screen through the built-in screen recording software of the system. The duration of the screen recorded video before the software to be analyzed runs abnormally can be 1 minute, 3 minutes, 5 minutes, 10 minutes and the like. In some embodiments, when an instruction of a user about abnormal operation of the software to be analyzed is obtained, recording is stopped, and a screen recorded video before the abnormal operation of the software to be analyzed is obtained. The screen interface can be a computer screen interface, a mobile phone screen interface, a vehicle-mounted screen interface and the like.
The screen recorded video before the software to be analyzed runs abnormally refers to a dynamic image recorded in an electric signal mode and consists of a plurality of continuous static images in time. Wherein each image is a frame of video data.
The running environment information of the software to be analyzed includes information of a plurality of running software in the device, operating system information, hardware device information, network configuration information, system setting information, and the like. The information of the plurality of running software in the device includes traffic usage data of the plurality of running software, memory occupation data of the plurality of running software, network connection data, and the like.
And S5, determining the activity degrees of the plurality of running software by using an activity degree determining model based on the screen recorded video before the running abnormality of the software to be analyzed and the running environment information of the software to be analyzed.
The screen recorded video before the abnormal operation of the software to be analyzed contains the operation conditions of a plurality of operation software before the abnormal operation of the software to be analyzed and the operation conditions of a user on the plurality of operation software, and whether the plurality of operation software are active or not can be judged through the screen recorded video before the abnormal operation of the software to be analyzed. For example, a screen recorded video shows that a user operates other running software while running the software to be analyzed, and opens a plurality of functional modules of the other running software, which means that the other running software is active in the device. If the running software uses more functional modules at runtime, the possibility of software conflicts may be increased to some extent.
The activity level of the running software may represent the running level of the running software of a plurality of functional modules during the running of the software to be analyzed. The greater the activity of the running software is, the more functional modules are run when the running software runs, and the running software can perform more operations when the equipment runs, so that the running software is more likely to collide with the software to be analyzed. As an example, when a user is shown in a screen recorded video before the software to be analyzed runs abnormally and runs a computer game, all functions of the game recording plug-in, such as shielding environmental sound, sound changing, multi-machine recording game and the like, are simultaneously opened, so that the recording plug-in has high activity in the game running, and the situation that the game recording plug-in occupies a sound channel of the game may occur, so that the user cannot use the voice function in the game process. As another example, in the operation process of the modem driver, when the user opens the antivirus software and opens the functions of antivirus operation, computer protection, network protection and the like in the screen recorded video before the software to be analyzed operates abnormally, the activity degree of the antivirus software is high, and the channel port of the modem driver may be occupied by the antivirus software, so that the situation that the operation of the modem driver is abnormal may occur.
The activity level determining model is a long-short-period neural network model. The long-term neural network model is one implementation of artificial intelligence. The Long and Short Term neural network model includes a Long and Short Term neural network (LSTM), which is one of RNNs (Recurrent Neural Network, recurrent neural networks). The long-term and short-term neural network model can process sequence data with any length, capture sequence information and output results based on the association relationship of front data and rear data in the sequence. The screen recorded video before the abnormal operation of the software to be analyzed in the continuous time period is processed through the long-short-term neural network model, and the characteristics of the association relationship between the screen recorded video before the abnormal operation of the software to be analyzed in each time point can be comprehensively considered can be output, so that the output characteristics are more accurate and comprehensive.
The input of the activity level determining model is recorded video of a screen before the running of the software to be analyzed is abnormal, and running environment information of the software to be analyzed, and the output of the activity level determining model is the activity level of the plurality of running software. The running environment information of the software to be analyzed can reflect the activity degree of the running software to a certain extent, for example, the higher the memory occupation of the running software is, the more frequent the network transmission is, and the higher the activity degree is, so that the running environment information of the software to be analyzed is also used as the input of an activity degree determination model to judge the activity degree of the running software, and the output result is more accurate.
The liveness determination model may be trained by training samples in the historical data. The training samples comprise sample input data and labels corresponding to the sample input data, the sample input data is a screen recorded video before the sample software to be analyzed runs abnormally and running environment information of the sample software to be analyzed, and the labels are the activity degrees of a plurality of running software of the sample. The output label of the training sample can be obtained through artificial labeling. For example, a user can judge a screen recorded video before the sample software to be analyzed runs abnormally and running environment information of the sample software to be analyzed, mark the activity degrees of a plurality of running software, and take the activity degrees of the marked running software as labels. In some embodiments, the initial liveness determination model may be trained by a gradient descent method to obtain a trained liveness determination model. Specifically, according to the training sample, constructing a loss function of the activity level determination model, adjusting the parameters of the activity level determination model through the loss function of the activity level determination model until the loss function value converges or is smaller than a preset threshold value, and completing training. The loss function may include, but is not limited to, a log (log) loss function, a square loss function, an exponential loss function, a range loss function, an absolute value loss function, and the like.
After training is completed, screen recorded video before the running abnormality of the software to be analyzed and running environment information of the software to be analyzed are input to an activity degree determination model after training is completed, and the activity degrees of the running software are obtained through output.
In some embodiments, the activity level determining model may further include a video processing model and an activity level output model, where the video processing model may process a screen recorded video before the abnormal operation of the software to be analyzed to determine a plurality of software running states in the video, and then determine the activity level of the plurality of running software using the activity level output model based on the plurality of software running states in the video and the running environment information of the software to be analyzed. The input of the video processing model is a screen recorded video before the running abnormality of the software to be analyzed, the output of the video processing model is a plurality of software running states in the video, the input of the activity degree output model is a plurality of software running states in the video and running environment information of the software to be analyzed, and the output of the activity degree output model is the activity degree of the plurality of running software.
The plurality of software running states represent running states of the plurality of software obtained by judging based on the screen recorded video before the software to be analyzed runs abnormally. The software running state may be a value between 0 and 1, with a larger value indicating that the software is running more actively.
The video processing model and the activity level output model are both long-term and short-term neural network models. The video processing model firstly identifies the running states of a plurality of pieces of software based on the screen recorded video before the running abnormality of the software to be analyzed, and then determines the activity degree of the plurality of pieces of running software according to the running environment information of the software to be analyzed and the running states of the plurality of pieces of software. The method has the advantages that the activity degree of a plurality of running software is determined through the sequential processing of the video processing model and the activity degree output model, and the problems that the data volume is too large, the loss function in training is difficult to converge, the training time is long and the like caused by the fact that only one model is used for simultaneously processing the screen recorded video before the running abnormality of the software to be analyzed and the running environment information of the software to be analyzed are avoided. The training time can be reduced and the training efficiency can be improved by using two models for processing.
And S6, determining one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed and the activity degree of the plurality of running software.
The associated software represents software which is associated with the software to be analyzed in a large degree in a plurality of running software and possibly incompatible with the software to be analyzed.
In some embodiments, the plurality of running software may be ordered according to the order of the interaction degree from big to small, the plurality of running software with the top N of the interaction degree ranking is selected, the plurality of running software is ordered according to the order of the activity degree from big to small, the plurality of running software with the top N of the activity degree ranking is selected, and finally the same software in the plurality of running software with the top N of the interaction degree ranking and the plurality of running software with the top N of the activity degree ranking is determined as one or more associated software, where N is a positive integer greater than or equal to 1. For example, N may be 5, 8, 10, etc. Fig. 2 is a schematic diagram of ordering a plurality of running software based on interaction level and activity level according to an embodiment of the present invention. As shown in fig. 2, the plurality of running software are respectively a browser, a antivirus software, a game, a mailbox and a chat software according to the order of the interaction degree from large to small, and the plurality of running software are respectively a player, a game, the antivirus software, the chat software and an input method according to the order of the activity degree from large to small, wherein the same software is the antivirus software, the game and the chat software, and the antivirus software, the game and the chat software are used as related software to carry out subsequent incompatibility judgment.
And S7, testing the one or more associated software in sequence to determine incompatible software corresponding to the software to be analyzed.
In some embodiments, the one or more associated software may be sequentially subjected to unloading operations, the software to be analyzed after each unloading operation is detected, and if it is detected that the software to be analyzed has no abnormal operation, the associated software corresponding to the unloading operation is used as incompatible software.
In some embodiments, the one or more associated software and the software to be analyzed may also be tested through a compatibility testing tool or a third party compatibility testing website, to test incompatible software corresponding to the software to be analyzed. The compatibility testing tool comprises a browser compatibility testing tool and an APP compatibility testing tool.
After the incompatible software is determined, the reason for abnormal operation of the software to be analyzed is found, and the incompatible software can be unloaded or modified later so as to avoid the occurrence of the incompatible condition again.
In some embodiments, if the incompatible software is not found after the sequential test, notifying a user to adopt other abnormal analysis means to analyze the software to be analyzed. The absence of incompatible software indicates that the software is not running abnormally due to incompatibility between the software. Other means of exception analysis include analyzing whether a software program component is missing, analyzing whether a system file is corrupted or lost, analyzing whether a hardware failure, etc.
Based on the same inventive concept, fig. 3 is a schematic diagram of an analysis system for software operation abnormality based on artificial intelligence according to an embodiment of the present invention, where the analysis system for software operation abnormality based on artificial intelligence includes: the receiving module 31 is configured to receive an instruction of abnormal operation of software to be analyzed sent by a user;
a first obtaining module 32, configured to obtain interaction data between the software to be analyzed and a plurality of running software;
an interaction degree determining module 33, configured to determine interaction degrees of the plurality of running software and the software to be analyzed using an interaction degree determining model based on interaction data of the software to be analyzed and the plurality of running software;
a second obtaining module 34, configured to obtain a screen recorded video before the software to be analyzed runs abnormally and running environment information of the software to be analyzed;
an activity level determining module 35, configured to determine activity levels of the plurality of running software by using an activity level determining model based on a screen recorded video before the running abnormality of the software to be analyzed and running environment information of the software to be analyzed;
an associated software determining module 36, configured to determine one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed, and the activity degree of the plurality of running software;
and the incompatible software determining module 37 is configured to sequentially test the one or more associated software and determine incompatible software corresponding to the software to be analyzed.
Based on the same inventive concept, an embodiment of the present invention provides an electronic device, as shown in fig. 4, including:
a processor 41; a memory 42 for storing executable program instructions in the processor 41; wherein the processor 41 is configured to execute to implement a method of analyzing software operating anomalies based on artificial intelligence as provided above, the method comprising: receiving an instruction of abnormal operation of software to be analyzed, which is sent by a user; acquiring interaction data of the software to be analyzed and a plurality of running software; determining the interaction degree of the plurality of running software and the software to be analyzed by using an interaction degree determining model based on the interaction data of the software to be analyzed and the plurality of running software; acquiring a screen recorded video before the operation abnormality of the software to be analyzed and the operation environment information of the software to be analyzed; determining the activity degrees of the plurality of running software by using an activity degree determining model based on the screen recorded video before the running abnormality of the software to be analyzed and the running environment information of the software to be analyzed; determining one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed and the activity degree of the plurality of running software; and sequentially testing the one or more associated software to determine incompatible software corresponding to the software to be analyzed.
Based on the same inventive concept, the present embodiment provides a non-transitory computer readable storage medium, which when executed by a processor 41 of an electronic device, enables the electronic device to perform an analysis method for implementing the artificial intelligence-based software operation anomaly as provided above, the method comprising receiving an instruction of the software operation anomaly to be analyzed issued by a user; acquiring interaction data of the software to be analyzed and a plurality of running software; determining the interaction degree of the plurality of running software and the software to be analyzed by using an interaction degree determining model based on the interaction data of the software to be analyzed and the plurality of running software; acquiring a screen recorded video before the operation abnormality of the software to be analyzed and the operation environment information of the software to be analyzed; determining the activity degrees of the plurality of running software by using an activity degree determining model based on the screen recorded video before the running abnormality of the software to be analyzed and the running environment information of the software to be analyzed; determining one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed and the activity degree of the plurality of running software; and sequentially testing the one or more associated software to determine incompatible software corresponding to the software to be analyzed.
Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments of this specification. Other variations are possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present specification may be considered as consistent with the teachings of the present specification. Accordingly, the embodiments of the present specification are not limited to only the embodiments explicitly described and depicted in the present specification.
Claims (8)
1. An artificial intelligence based analysis method for software operation abnormality is characterized by comprising the following steps:
receiving an instruction of abnormal operation of software to be analyzed, which is sent by a user;
acquiring interaction data of the software to be analyzed and a plurality of running software;
determining the interaction degree of the plurality of running software and the software to be analyzed by using an interaction degree determining model based on the interaction data of the software to be analyzed and the plurality of running software;
acquiring a screen recorded video before the operation abnormality of the software to be analyzed and the operation environment information of the software to be analyzed;
determining the activity degrees of the plurality of running software by using an activity degree determining model based on the screen recorded video before the running abnormality of the software to be analyzed and the running environment information of the software to be analyzed;
determining one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed and the activity degree of the plurality of running software;
and sequentially testing the one or more associated software to determine incompatible software corresponding to the software to be analyzed.
2. The method for analyzing abnormal software operation based on artificial intelligence according to claim 1, wherein the interaction degree determining model is a deep neural network model, the input of the interaction degree determining model is interaction data of the software to be analyzed and a plurality of running software, and the output of the interaction degree determining model is interaction degree of the plurality of running software and the software to be analyzed.
3. The method for analyzing abnormal software operation based on artificial intelligence according to claim 1, wherein the activity level determining model is a long-short-term neural network model, the input of the activity level determining model is a screen recorded video before the abnormal software operation to be analyzed and the operation environment information of the software operation to be analyzed, and the output of the activity level determining model is the activity level of the plurality of operation software.
4. The method for analyzing abnormal operation of software based on artificial intelligence according to claim 1, wherein the determining of one or more associated software is based on the interaction degree of the plurality of running software with the software to be analyzed and the activity degree of the plurality of running software; comprising the following steps:
sequencing the plurality of running software according to the sequence of the interaction degree from large to small, and selecting a plurality of running software with the top N of the interaction degree ranking;
sequencing the plurality of running software according to the sequence of the activity degrees from large to small, and selecting a plurality of running software with the activity degrees ranked N at the top;
and determining the same software in the running software with the interaction degree of N ranked in front and the running software with the activity degree of N ranked in front as one or more associated software, wherein N is a positive integer greater than or equal to 1.
5. The method for analyzing abnormal operation of software based on artificial intelligence according to claim 1, wherein the sequentially testing the one or more associated software to determine incompatible software corresponding to the software to be analyzed comprises: and carrying out unloading operation on the one or more associated software in sequence, detecting the software to be analyzed after each unloading operation, and taking the associated software corresponding to the unloading operation as incompatible software if the fact that the software to be analyzed is not abnormal in operation is detected.
6. The method for analyzing software operation anomalies based on artificial intelligence according to claim 5, further comprising: and if the incompatible software is not found after the sequential test, notifying a user to adopt other abnormal analysis means to analyze the software to be analyzed.
7. An artificial intelligence based analysis system for software operational anomalies, comprising:
the receiving module is used for receiving an instruction of abnormal operation of the software to be analyzed, which is sent by a user;
the first acquisition module is used for acquiring interaction data of the software to be analyzed and a plurality of running software;
the interaction degree determining module is used for determining interaction degrees of the plurality of running software and the software to be analyzed by using an interaction degree determining model based on interaction data of the software to be analyzed and the plurality of running software;
the second acquisition module is used for acquiring a screen recorded video before the operation abnormality of the software to be analyzed and the operation environment information of the software to be analyzed;
the activity degree determining module is used for determining the activity degrees of the plurality of running software by using an activity degree determining model based on the screen recorded video before the running abnormality of the software to be analyzed and the running environment information of the software to be analyzed;
the associated software determining module is used for determining one or more associated software based on the interaction degree of the plurality of running software and the software to be analyzed and the activity degree of the plurality of running software;
and the incompatible software determining module is used for sequentially testing the one or more associated software and determining incompatible software corresponding to the software to be analyzed.
8. The artificial intelligence based software anomaly analysis system of claim 7, wherein the associated software determination module is further configured to:
sequencing the plurality of running software according to the sequence of the interaction degree from large to small, and selecting a plurality of running software with the top N of the interaction degree ranking;
sequencing the plurality of running software according to the sequence of the activity degrees from large to small, and selecting a plurality of running software with the activity degrees ranked N at the top;
and determining the same software in the running software with the interaction degree of N ranked in front and the running software with the activity degree of N ranked in front as one or more associated software, wherein N is a positive integer greater than or equal to 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310282473.4A CN115994098B (en) | 2023-03-22 | 2023-03-22 | Analysis method and system for software operation abnormality based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310282473.4A CN115994098B (en) | 2023-03-22 | 2023-03-22 | Analysis method and system for software operation abnormality based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115994098A true CN115994098A (en) | 2023-04-21 |
| CN115994098B CN115994098B (en) | 2024-01-16 |
Family
ID=85992387
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310282473.4A Active CN115994098B (en) | 2023-03-22 | 2023-03-22 | Analysis method and system for software operation abnormality based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115994098B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102013210160A1 (en) * | 2012-06-15 | 2013-12-19 | International Business Machines Corporation | Method for facilitating processing of transaction within multiprocessor data processing environment, involves automatically setting counter to predefined value based on presenting interrupt of transaction when transaction is aborted |
| CN105827455A (en) * | 2016-04-27 | 2016-08-03 | 乐视控股（北京）有限公司 | Method and apparatus for modifying resource model |
| CN109213675A (en) * | 2017-07-05 | 2019-01-15 | 瞻博网络公司 | Software analysis platform |
| CN109376071A (en) * | 2018-09-03 | 2019-02-22 | 平安普惠企业管理有限公司 | Application software exception feedback method, device, computer equipment and storage medium |
| CN109408374A (en) * | 2018-09-26 | 2019-03-01 | 平安普惠企业管理有限公司 | Application exception is remotely located method, system, computer equipment and medium |
| CN109901831A (en) * | 2019-01-22 | 2019-06-18 | 成都随安保网络技术有限公司 | The multi-platform compatibility operation method and compatibility operation device of software |
| CN112764784A (en) * | 2021-02-03 | 2021-05-07 | 河南工业大学 | Automatic software defect repairing method and device based on neural machine translation |
| CN113495841A (en) * | 2021-06-28 | 2021-10-12 | 北京百度网讯科技有限公司 | Compatibility detection method, device, equipment, storage medium and program product |
| US20210326095A1 (en) * | 2019-08-30 | 2021-10-21 | Capital One Services, Llc | Application replication platform |
| CN113704081A (en) * | 2020-10-15 | 2021-11-26 | 天翼智慧家庭科技有限公司 | Automatic testing method and system for application program compatibility |
| CN114464180A (en) * | 2022-02-21 | 2022-05-10 | 海信电子科技(武汉)有限公司 | Intelligent device and intelligent voice interaction method |
| CN114936153A (en) * | 2022-05-17 | 2022-08-23 | 青岛百洋智能科技股份有限公司 | Turing test method of artificial intelligence software |
-
2023
- 2023-03-22 CN CN202310282473.4A patent/CN115994098B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102013210160A1 (en) * | 2012-06-15 | 2013-12-19 | International Business Machines Corporation | Method for facilitating processing of transaction within multiprocessor data processing environment, involves automatically setting counter to predefined value based on presenting interrupt of transaction when transaction is aborted |
| CN105827455A (en) * | 2016-04-27 | 2016-08-03 | 乐视控股（北京）有限公司 | Method and apparatus for modifying resource model |
| CN109213675A (en) * | 2017-07-05 | 2019-01-15 | 瞻博网络公司 | Software analysis platform |
| CN109376071A (en) * | 2018-09-03 | 2019-02-22 | 平安普惠企业管理有限公司 | Application software exception feedback method, device, computer equipment and storage medium |
| CN109408374A (en) * | 2018-09-26 | 2019-03-01 | 平安普惠企业管理有限公司 | Application exception is remotely located method, system, computer equipment and medium |
| CN109901831A (en) * | 2019-01-22 | 2019-06-18 | 成都随安保网络技术有限公司 | The multi-platform compatibility operation method and compatibility operation device of software |
| US20210326095A1 (en) * | 2019-08-30 | 2021-10-21 | Capital One Services, Llc | Application replication platform |
| CN113704081A (en) * | 2020-10-15 | 2021-11-26 | 天翼智慧家庭科技有限公司 | Automatic testing method and system for application program compatibility |
| CN112764784A (en) * | 2021-02-03 | 2021-05-07 | 河南工业大学 | Automatic software defect repairing method and device based on neural machine translation |
| CN113495841A (en) * | 2021-06-28 | 2021-10-12 | 北京百度网讯科技有限公司 | Compatibility detection method, device, equipment, storage medium and program product |
| CN114464180A (en) * | 2022-02-21 | 2022-05-10 | 海信电子科技(武汉)有限公司 | Intelligent device and intelligent voice interaction method |
| CN114936153A (en) * | 2022-05-17 | 2022-08-23 | 青岛百洋智能科技股份有限公司 | Turing test method of artificial intelligence software |
Non-Patent Citations (2)
| Title |
|---|
| ORI LAHAV 等: "Pruning and Slicing Neural Networks using Formal Verification", 《2021 FORMAL METHODS IN COMPUTER AIDED DESIGN》 * |
| 管华 等: "面向服务软件异常处理研究综述", 《计算机科学》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115994098B (en) | 2024-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107294808B (en) | Interface test method, device and system | |
| CN111897724B (en) | Automatic testing method and device suitable for cloud platform | |
| CN111124919A (en) | User interface testing method, device, equipment and storage medium | |
| US20070022324A1 (en) | Multi-platform test automation enhancement | |
| CN111552633A (en) | Interface abnormal call testing method and device, computer equipment and storage medium | |
| CN108241580B (en) | Client program testing method and terminal | |
| CN114844768B (en) | Information analysis method and device and electronic equipment | |
| CN108228444B (en) | Test method and device | |
| CN110784374A (en) | Method, device, equipment and system for monitoring operation state of service system | |
| CN111522749B (en) | Page testing method and device, readable storage medium and electronic equipment | |
| US20180357143A1 (en) | Testing computing devices | |
| CN111198797B (en) | Operation monitoring method and device and operation analysis method and device | |
| CN109409948B (en) | Transaction abnormity detection method, device, equipment and computer readable storage medium | |
| CN115994098B (en) | Analysis method and system for software operation abnormality based on artificial intelligence | |
| CN112416735B (en) | Application program detection method and device, terminal equipment and storage medium | |
| CN117294833A (en) | Camera chip testing method and related equipment | |
| CN117041111A (en) | Vehicle cloud function test method and device, electronic equipment and storage medium | |
| CN116166536A (en) | Test method, test device, electronic equipment and storage medium | |
| CN116260643A (en) | Security testing method, device and equipment for web service of Internet of things | |
| CN111026631B (en) | Automatic interface detection method, device and server | |
| JP2024526023A (en) | A self-optimizing analysis system for core dumps | |
| CN115866235A (en) | Video quality evaluation method and device, electronic equipment and storage medium | |
| CN111526041A (en) | Test method and device for centralized control platform | |
| CN113849364B (en) | Edge application management method, device, equipment and readable storage medium | |
| CN113407440B (en) | Testing system and method for wireless communication module |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20231226 Address after: Chinatelecom tower, No. 19, Chaoyangmen North Street, Dongcheng District, Beijing 100010 Applicant after: Tianyi Safety Technology Co.,Ltd. Address before: No. 324 Huixiang Road, Xinfan Street, Xindu District, Chengdu City, Sichuan Province, 610000 Applicant before: Chengdu Daqianyan Software Development Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |