Defect reason determination method and device, electronic equipment, storage medium and product

Abstract

The invention discloses a defect reason determination method and device, electronic equipment, a storage medium and a product, and relates to the technical field of artificial intelligence. The defect cause determination method comprises the following steps: determining sample defect reports matched with at least two sample objects; for each sample object, according to at least two preset defect reason types and a sample defect report matched with the current sample object, determining a prompt text matched with the current sample object; constructing a current sample data pair according to the prompt text matched with the current sample object and the defect reason type of the current sample object; and training the large language model according to the sample data pairs corresponding to the sample objects to obtain a defect reason classification model, and determining defect reasons according to the defect reason classification model. According to the technical scheme, defect reason classification is carried out based on the defect classification reason model obtained through large language model training, and the accuracy and objectivity of defect reason analysis can be improved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of artificial intelligence technology, and in particular, to a method, device, electronic device, storage medium and product for determining the cause of a defect. Background Art

[0002] During the development, testing and implementation of software, systems or projects, defect analysis is the process of in-depth research and evaluation of deficiencies or problems in software, systems or projects. Defect management tools provide powerful defect tracking and analysis functions, which can not only realize the status tracking of individual defects (such as the status flow of defect creation, location, resolution, closure, etc., as well as the location of defect causes and processing comments of each defect processing subject), but also realize the analysis of overall defect trends.

[0003] However, when defect tracking is performed through defect management tools, the defect status circulates among all defect handling entities related to the defect. There may be differences in the defect cause location and handling comments of different defect handling entities, or there may be misoperations, resulting in unclear final defect cause analysis results. Summary of the invention

[0004] The embodiments of the present invention provide a method, device, electronic device, storage medium and product for determining the cause of a defect, so as to solve the obstacle of human interference factors in tracing the cause of the defect.

[0005] In a first aspect, an embodiment of the present invention provides a method for determining a defect cause, comprising:

[0006] identifying sample defect reports that match at least two sample objects;

[0007] For each sample object, according to at least two preset defect cause types and a sample defect report matching the current sample object, determine a prompt text matching the current sample object;

[0008] Constructing a current sample data pair according to the prompt text matching the current sample object and the defect cause type of the current sample object;

[0009] The large language model is trained according to the sample data pairs corresponding to each sample object to obtain a defect cause classification model, so as to determine the defect cause according to the defect cause classification model.

[0010] In a second aspect, an embodiment of the present invention provides a method for determining a defect cause, including:

[0011] Determine the defect report that matches the object to be processed;

[0012] Determine a prompt text matching the object to be processed according to at least two preset defect cause types and the defect report matching the object to be processed;

[0013] Inputting the prompt text matching the object to be processed into a defect cause classification model to obtain the defect cause type of the object to be processed output by the defect cause classification model;

[0014] Wherein, the defect cause classification model is trained using the method described in any one of the above embodiments.

[0015] In a third aspect, an embodiment of the present invention further provides a device for determining a defect cause, the device comprising:

[0016] A sample defect report determination module, used to determine a sample defect report that matches at least two sample objects;

[0017] A current sample object prompt text determination module is used to determine, for each sample object, a prompt text matching the current sample object according to at least two preset defect cause types and a sample defect report matching the current sample object;

[0018] A current sample data pair determination module, used to construct a current sample data pair according to a prompt text matched with a current sample object and a defect cause type of the current sample object;

[0019] The defect cause classification model determination module is used to train the large language model according to the sample data pairs corresponding to each sample object to obtain the defect cause classification model, so as to determine the defect cause according to the defect cause classification model.

[0020] In a fourth aspect, an embodiment of the present invention further provides a device for determining a defect cause, the device comprising:

[0021] A defect report determination module for an object to be processed, used to determine a defect report matching the object to be processed;

[0022] A prompt text determination module for an object to be processed, used to determine a prompt text matching the object to be processed according to at least two preset defect cause types and the defect report matching the object to be processed;

[0023] A module for determining the defect cause type of the object to be processed, used for inputting the prompt text matching the object to be processed into the defect cause classification model to obtain the defect cause type of the object to be processed output by the defect cause classification model;

[0024] Wherein, the defect cause classification model is trained using the method described in any one of the above embodiments.

[0025] In a fifth aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for determining the cause of a defect as described in any one of the embodiments of the present invention is implemented.

[0026] In a sixth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for determining a defect cause as described in any one of the embodiments of the present invention.

[0027] In a seventh aspect, an embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements a method for determining a defect cause as described in any one of the embodiments of the present invention.

[0028] In the embodiment of the present invention, by determining the sample defect report of each sample object, a prompt text is generated for the current sample object according to the preset defect cause type and the sample defect report of the current sample object, a current sample data pair is constructed according to the prompt text of the current sample object and its defect cause type, and a large language model is trained according to the sample data pairs corresponding to each sample object to obtain a defect cause classification model, so as to determine the defect cause according to the defect cause classification model. The problem of inconsistent defect cause positioning when defect tracking is performed through defect management tools in the prior art is solved, the defect cause classification model is trained based on the large model, the actual cause of the defect is comprehensively judged, the accuracy and rationality of defect cause classification are improved, and the obstacles to tracking the actual defect cause caused by human interference factors are reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0030] Figure 1 A flowchart of a method for determining a defect cause provided by an embodiment of the present invention;

[0031] Figure 2 A flowchart of a method for determining a defect cause provided by an embodiment of the present invention;

[0032] Figure 3 A schematic diagram of the structure of a device for determining the cause of a defect provided by an embodiment of the present invention;

[0033] Figure 4A schematic diagram of the structure of a device for determining the cause of a defect provided by an embodiment of the present invention;

[0034] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0036] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance. The acquisition, storage, use, processing, etc. of data in the technical solution of this application comply with the relevant provisions of national laws and regulations. It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. The purpose is only to illustrate the feasibility of the implementation of the technical solution of this application, but it does not mean that the applicant has or must use the relevant content of the solution.

[0037] Figure 1 A flowchart of a method for determining a defect cause provided by an embodiment of the present invention, the embodiment of the present invention can be applied to the case where a defect cause classification model is trained based on a large language model for defect cause analysis. The embodiment of the present invention can be implemented by a device for determining the defect cause, the device for determining the defect cause can be implemented in the form of hardware and / or software, and the device can be deployed in a server.

[0038] like Figure 1 As shown, the method for determining the cause of a defect in an embodiment of the present invention includes:

[0039] S110: Determine a sample defect report that matches at least two sample objects.

[0040] This embodiment provides a technical solution for training a defect cause classification model based on a large language model.

[0041] The sample object is an object that needs to be analyzed and located for defect causes, and may be software, a system, or a project, etc. This embodiment does not limit the type of the sample object.

[0042] A sample defect report refers to an analysis report obtained by tracking defects on a sample object. The sample defect report may include detailed defect information of the sample object (such as defect description information, screenshots, related code snippets, etc.), defect status (such as new creation, assignment, processing, repair, verification, and closure, etc.), the flow process of the defect status throughout its life cycle, and the final defect solution.

[0043] In this embodiment, the sample defect report matching the sample object can be obtained through a defect management tool, which is a collection of tools and processes for managing and monitoring defects in software, systems, or products.

[0044] Furthermore, the defect management tool can be used to track defects on sample objects to obtain a sample defect report; in the defect management tool, a sample defect report matching the sample object is obtained. The defect management tool is used to track defects on sample objects to obtain a sample defect report, which can specifically include: if it is determined that the sample object has a defect, a defect is created in the defect management tool; the created defect is assigned to the defect handling subject, and the defect handling measures uploaded by the defect handling subject are determined; the defect handling measures are verified; if it is determined that the defect handling measures are verified, the defect is closed, otherwise, the execution is returned to assign the created defect to the defect handling subject until the defect handling measures are verified; the defect management tool generates a sample defect report based on the defect description information, defect status, defect flow process, and defect handling measures before the defect is closed during the process from defect creation to closure.

[0045] It should be noted that multiple defects may occur during the development, testing and implementation of the sample object. In this case, each defect needs to be tracked separately, the corresponding sample defect report needs to be determined, and then subsequent prompt text generation and sample data pair generation can be performed separately.

[0046] S120: For each sample object, determine a prompt text matching the current sample object according to at least two preset defect cause types and a sample defect report matching the current sample object.

[0047] Among them, the defect cause type is used to indicate the cause of the defect. For example, taking the sample object as software, the defect cause type can be set to the following items: front-end logic problem, back-end logic problem, front-end display problem, front-end and back-end collaboration problem, test environment / test data problem, tester understanding problem, unclear requirements / change problem and other problems. However, this embodiment does not limit the number and specific content of the pre-set defect cause types for different types of sample objects.

[0048] The prompt text is also called a prompt (hereinafter referred to as a prompt text), which is the initial text input received by the large language model to generate a response or complete a task. The prompt can be in the form of a question, a description, or a group of keywords. In this embodiment, the prompt is in the form of text data.

[0049] The prompt text in this embodiment may include instruction information, input information, background information, and output instructions, etc. The instruction information is used to inform the large language model of the specific task to be performed; the input information is the specific information required by the large language model when performing the task; the background information provides the large language model with the environment information or additional details required to perform the task; the output instruction is used to indicate the format or type of data output by the large language model.

[0050] In this embodiment, different defect cause types are set in advance, and based on the pre-set defect cause types and sample defect reports, a structured prompt text corresponding to the current sample object is generated. This setting enables the large language model to ultimately attribute the defect of the current sample object to one of the defect cause types during the execution of the defect cause classification task. At the same time, the prompt text generated based on the sample defect report can retain the original information of the defect of the current sample object to the greatest extent possible, and construct a comprehensive defect representation, thereby making the final defect cause type positioning more accurate and reasonable.

[0051] In an optional embodiment, further, S120 may include:

[0052] A1. Using at least two pre-set defect cause types as background information of the prompt text matching the current sample object;

[0053] A2. Determine the defect description information and defect flow information of the current sample object according to the sample defect report matching the current sample object;

[0054] The defect flow information is a record of the state changes of the defect of the current sample object in the entire life cycle and at least one defect handling process;

[0055] A3. Use the defect description information and defect flow information of the current sample object as input information of the prompt text matching the current sample object.

[0056] In this embodiment, each defect cause type is used as the background information part of the prompt text, which can enable the large language model to better understand the contextual information and constraints in the defect cause classification task, so that the large language model can determine the defect cause type of the current sample object among various defect cause types, thereby ensuring the clarity and effectiveness of the defect cause classification task.

[0057] In this embodiment, the sample defect report includes description information and flow information of the defect of the current sample object. Defect description information refers to relevant information that describes the defect of the current sample object in detail, accurately and clearly, and may include defect title, defect detailed description, reproduction steps, expected results, actual results, and environmental information. Among them, the defect title summarizes the main problems of the defect in concise and clear language; the defect detailed description provides a detailed description of the specific manifestations of the defect, including under what circumstances the defect occurs, what the specific error condition is, whether there is an error prompt message, etc.; the reproduction steps provide a specific operation process for stably reproducing the defect; the expected result is used to describe the result that should appear under normal circumstances; the actual result is the actual result that does not meet the expected result; the environmental information provides relevant environmental information when the defect occurs, such as the operating system, browser type and version, and software version.

[0058] In a specific example, taking the current sample object as a certain software, the defect title can be "The login page cannot submit the form", the defect detailed description can be "After entering the correct username and password, click the login button, the page does not respond at all, and no error prompt appears", the reproduction steps can be "1. Open the login page; 2. Enter the username 'testuser' and password '123456'; 3. Click the login button", the expected result can be "Should successfully jump to the user's personal homepage"; the actual result can be "The page remains unchanged, without any jump or prompt"; the environmental information can be "xx operating system, xx browser version".

[0059] Defect flow information is a detailed record of the state changes of the defects of the current sample object during the entire life cycle and at least one defect handling process. Defect flow information may include the defect creation time, defect status, defect handling subject, defect handling time, defect handling description, and defect handling result of the current sample object.

[0060] In a specific example, taking the current sample object as a certain software, the defect flow information may include the following content: defect creation time: xx / xx / xx / xx; discoverer: tester A; current status: verified; historical status records: newly created at t1, assigned to developer B at t2, being processed at t3, fixed at t4, verified at t5; handler: developer B; processing time: t3-t4; processing description: checking the code and finding that the database connection parameters were wrong, the defect was fixed after modification and retesting; verifier: tester A; verification time: t5; verification result: passed verification, the defect has been fixed.

[0061] Furthermore, when there are multiple defect handling processes, it may also include a process for transferring defect status, defect handling subjects, etc. between multiple defect handling processes; when there are multiple defect handling subjects, it may also include a process for defect transfer between multiple defect handling subjects, which will not be repeated in this embodiment.

[0062] In this embodiment, the sample defect report includes multi-dimensional defect-related data such as the defect description information and defect flow information of the current sample object. By determining the input information of the prompt text of the current sample object based on the sample defect report, a comprehensive defect representation can be constructed, so that the large language model can fully understand the specific defect information of the current sample object when classifying the defect cause of the current sample object, thereby improving the accuracy and rationality of the defect cause analysis and reducing the interference of human factors on the location of the actual cause of the defect.

[0063] Furthermore, when determining the input information of the prompt text of the current sample object based on the defect description information and defect flow information in the sample defect report, the rich text information in the sample defect report, such as images, fonts, font sizes, etc., can be removed to retain only the plain text information. This ensures that the prompt text of the current sample object is plain text information, thereby ensuring that the input information of the large language model is in the correct format.

[0064] Furthermore, A3 may include:

[0065] A31. Determine at least two defect handling subject identifiers in the defect flow information of the current sample object;

[0066] A32. Add the role information of the defect handling subject to each defect handling subject identification;

[0067] A33. The defect flow information of the current sample object after adding the defect processing subject role information is used as input information of the prompt text matching the current sample object.

[0068] In this embodiment, the defect flow information of the sample defect report of the current sample object usually includes the identification of the defect handling subject, but lacks the role information of the defect handling subject. Therefore, this embodiment adds the role information corresponding to each defect handling subject identification to the defect flow information of the sample defect report of the current sample object.

[0069] In this embodiment, after determining the defect handling subject identification of the defect flow information of the sample defect report, adding the defect handling subject role information is helpful to assist in locating the cause of the defect. Locating the cause of the defect in combination with the role information of the handling subject in each stage of defect handling can make the defect cause analysis more credible and objective.

[0070] Furthermore, in this embodiment, a mapping relationship between each defect handling subject identifier and its corresponding defect handling subject role information can be pre-stored. After determining the defect handling subject identifier in the defect flow information of the current sample object, the corresponding role information is determined according to the mapping relationship.

[0071] In a specific example, the defect transfer information of the sample defect report may include the following content: newly created by zhangsan at t1; transferred by lisi to wangwu at t2, with comments: the back-end interface is abnormal, not a front-end problem; changed the status to located by wangwu at t3, with comments: the value is wrong, and it does not match the requirement document; changed the status to pending verification by wangwu at t4, with comments: the requirements personnel have been communicated and confirmed, the existing requirements have been adjusted, and the back-end has been updated accordingly; closed by zhangsan at t5. Zhangsan, Lisi and Wangwu are all defect handling subject identifiers. The defect flow information after adding role information to them includes: newly created by Zhangsan (tester) at t1; transferred from Lisi (front-end developer) to Wangwu (back-end developer) at t2, with comments: the back-end interface is abnormal, not a front-end problem; changed the status to Located by Wangwu (back-end developer) at t3, with comments: the value is wrong, inconsistent with the requirement document; changed the status to To be verified by Wangwu (back-end developer) at t4, with comments: the requirement personnel have been communicated and confirmed, the existing requirements have been adjusted, and the back-end has been updated accordingly; closed by Zhangsan (tester) at t5.

[0072] Furthermore, A3 may include:

[0073] A301. Determine, according to defect flow information of the current sample object, timing information of at least two defect handling processes of the current sample object;

[0074] A302, adding timing information to each defect processing process of the current sample object;

[0075] A303. Use the defect flow information of the current sample object after adding the timing information as input information of the prompt text matching the current sample object.

[0076] The timing information refers to the duration of the defect of the current sample object in each defect processing process, and can be calculated based on the start time of the two previous and subsequent defect processing processes.

[0077] In this embodiment, the start time of each defect handling process is extracted, and the duration of each defect handling process is calculated, the timing information of each defect handling process is added to the defect flow information of the current sample object, the time dimension is introduced in the defect flow information, and the input information of the prompt text of the current sample object is determined according to the defect flow information after adding the timing information, so that the large language model can capture the time pattern in the defect tracking process. At the same time, when processing multiple sample objects or multiple defects, there is no need to consume computing resources to flatten the timing data, which improves the model training efficiency and defect cause analysis efficiency.

[0078] In a specific example, still taking the defect flow information of the current sample object as an example, on the basis of adding the defect handling subject role information, the defect flow information after further adding the defect handling process timing information includes: newly created by zhangsan (tester) at time t1; transferred by lisi (front-end developer) to wangwu (back-end developer) at time t2, processing comments: back-end interface exception, not front-end problem, the previous stage took t2-t1; wangwu (back-end developer) changed the status to located at time t3, processing comments: value error, inconsistent with the requirement document, the previous stage took t3-t2; wangwu (back-end developer) changed the status to pending verification at time t4, and transferred it to zhangsan (tester), processing comments: communicated and confirmed with the requirement personnel, adjusted the existing requirements, and the back-end has been updated accordingly, the previous stage took t4-t3; closed by zhangsan (tester) at time t5, the previous stage took t5-t4.

[0079] Furthermore, the prompt text of the current sample object also includes instruction information; the instruction information is used to instruct the large language model to determine the defect cause type of the current sample object from various defect cause types of the background information according to the input information.

[0080] The instruction information is an essential component of the prompt text of the large language model, and is used to inform the large language model of the specific tasks to be performed. In this embodiment, the instruction information is used to inform the large language model to determine the defect cause type of the current sample object from the various defect cause types provided in the background information based on the input information.

[0081] The prompt text of the current sample object may further include a large language model role, which is used to indicate the role played by the large language model in performing a defect cause classification task.

[0082] In a specific example, taking the current sample object as software, this embodiment provides an example of the prompt text of the current sample object: You are a software development manager (large language model role), and you need to analyze the actual cause of the defect (instruction information) from the following defect description and processing process. The cause classification includes: back-end logic problems, front-end logic problems, front-end display problems, front-end and back-end collaboration problems, test environment / test data problems, tester understanding problems, unclear requirements / changes, and other problems (background information). The following is the defect description information: Open the xx page and click the xx menu. Click the xx button to pop up an error message "Failed to process the request". The following is the defect flow information: created by zhangsan (tester) at t1; transferred from lisi (front-end developer) to wangwu (back-end developer) at t2, processing comments: back-end interface abnormality, not front-end problem, the previous stage took t2-t1; wangwu (back-end developer) changed the status to located at t3, processing comments: value error, inconsistent with the requirements document, the previous stage took t3-t2; wangwu (back-end developer) changed the status to pending verification at t4, transferred to zhangsan (tester), processing comments: confirmed with the requirements personnel, adjusted the existing requirements, the back-end has been updated accordingly, the previous stage took t4-t3; closed by zhangsan (tester) at t5, the previous stage took t5-t4 (input information). Please analyze the cause type of the defect and give detailed analysis results (output instructions).

[0083] Furthermore, when generating the prompt text matching the current sample object, the defect flow information of the sample defect report of the current sample object may be subjected to data enhancement processing. Specifically, the data enhancement processing may include extracting keywords, replacing keywords with synonyms, and / or reorganizing the corpus. In this embodiment, the data enhancement processing may enrich the sample data and improve the understanding and generalization capabilities of the defect cause classification model.

[0084] In another optional embodiment, further, A3 may include: extracting keywords from the defect flow information of the current sample object to obtain at least one keyword; and determining the input information of the prompt text matching the current sample object based on the extracted keywords and the defect description information of the current sample object.

[0085] On the basis of the above-mentioned embodiment, this embodiment provides a technical solution for extracting keywords from defect flow information and then generating a prompt text.

[0086] According to the example of the prompt text of the current sample object mentioned above, the defect flow information occupies a large space, which retains more comprehensive defect-related information and can make the results of the defect cause analysis more accurate and reasonable. On this basis, the defect flow information is extracted with keywords, for example, the defect status, role information of the defect handling subject, defect handling description, time consumption and other information in each defect handling process can be extracted, and only the keywords are retained to generate the prompt text. This can improve the efficiency of defect cause analysis while retaining as much defect-related information as possible. At the same time, when the large language model is trained through the prompt text of each sample object, the training efficiency can be improved.

[0087] Still taking the prompt text of the above-mentioned current sample object as an example, after keyword extraction, only the keywords retaining the defect flow information may include: newly created by tester; transferred by front-end developer to back-end developer, processing comments: back-end interface abnormality, not front-end problem, the previous stage took t2-t1; back-end developer changes the status to located, processing comments: value error, inconsistent with the requirement document, the previous stage took t3-t2; back-end developer changes the status to pending verification, transfers to tester, processing comments: communicated and confirmed with the requirement personnel, adjusted the existing requirements, and the back-end has made corresponding updates, the previous stage took t4-t3; closed by tester, the previous stage took t5-t4.

[0088] S130: Construct a current sample data pair according to the prompt text matching the current sample object and the defect cause type of the current sample object.

[0089] In this embodiment, after determining the prompt text matching the current sample object, <prompt text, defect cause type> is used as a current sample data pair. For each sample object, the method of S120-S130 is adopted to construct its sample data pair, and finally a set of sample data pairs corresponding to each sample object is obtained for large model training.

[0090] In this embodiment, the defect cause type of the current sample object can be determined based on the sample defect report of the current sample object, or the defect cause type of each sample object can be marked in advance. This embodiment does not limit the method for determining the defect cause type of the sample object.

[0091] S140. Train the large language model according to the sample data pairs corresponding to each sample object to obtain a defect cause classification model, so as to determine the defect cause according to the defect cause classification model.

[0092] The large language model can learn complex patterns and features by training massive data, and has strong generalization, expression and prediction capabilities. In this embodiment, the set of sample data pairs corresponding to each sample object is used as training data, and the large language model is fine-tuned to obtain a defect cause classification model. Based on the large model training defect cause classification model, the actual cause of the defect is comprehensively judged, and all information can be retained to the maximum extent, so that the defect cause classification can be more accurate and logical, and the obstacles to tracking the actual cause of the defect caused by human interference factors can be reduced, and problems in the development, testing or implementation of the object to be processed can be discovered in a timely manner.

[0093] Furthermore, in the process of fine-tuning the large language model according to the set of sample data pairs corresponding to each sample object, the defect cause type prediction result of the large language model is compared with the actual defect cause type of each sample object. When the prediction accuracy of the large language model is greater than or equal to the preset accuracy threshold, the fine-tuning training is stopped to obtain the defect cause classification model.

[0094] In the embodiment of the present invention, by determining the sample defect report of each sample object, a prompt text is generated for the current sample object according to the preset defect cause type and the sample defect report of the current sample object, a current sample data pair is constructed according to the prompt text of the current sample object and its defect cause type, and a large language model is trained according to the sample data pairs corresponding to each sample object to obtain a defect cause classification model, so as to determine the defect cause according to the defect cause classification model. The problem of inconsistent defect cause positioning when defect tracking is performed through defect management tools in the prior art is solved, the defect cause classification model is trained based on the large model, the actual cause of the defect is comprehensively judged, the accuracy and rationality of defect cause classification are improved, and the obstacles to tracking the actual defect cause caused by human interference factors are reduced.

[0095] Figure 2 A flowchart of a method for determining the cause of a defect provided by an embodiment of the present invention is applicable to the case where the cause of a defect is analyzed during the development, testing or implementation of an object to be processed. The embodiment of the present invention can be implemented by a device for determining the cause of a defect, which can be implemented in the form of hardware and / or software, and can be deployed in a server.

[0096] like Figure 2 As shown, the method for determining the cause of a defect in an embodiment of the present invention includes:

[0097] S210: Determine a defect report that matches the object to be processed.

[0098] The process of determining the defect report of the object to be processed is the same as the process of determining the sample defect report of the current sample object, and will not be described in detail in this embodiment.

[0099] It should be noted that the object to be processed must be consistent with the type of sample object used to train the defect cause classification model.

[0100] S220: Determine a prompt text matching the object to be processed according to at least two preset defect cause types and the defect report matching the object to be processed.

[0101] Similarly, the process of generating a prompt text that matches the object to be processed based on the defect cause type and the defect report that matches the object to be processed is the same as the process of generating a prompt text that matches the current sample object based on the defect cause type and the sample defect report that matches the current sample object, and this embodiment will not be repeated here.

[0102] Furthermore, S220 may include:

[0103] B1. Using at least two pre-set defect cause types as background information of the prompt text matching the object to be processed;

[0104] B2. Determine defect description information and defect flow information of the object to be processed according to the defect report matching the object to be processed;

[0105] The defect flow information is a record of the state changes of the defects of the object to be processed during the entire life cycle and at least one defect processing process;

[0106] B3. Use the defect description information and defect flow information of the object to be processed as input information of the prompt text matching the object to be processed.

[0107] Furthermore, B3 may include:

[0108] B31. Determine at least two defect handling subject identifiers in the defect transfer information of the object to be handled;

[0109] B32. Add the role information of the defect handling subject to each defect handling subject identification;

[0110] B33. The defect flow information of the object to be processed after adding the defect processing subject role information is used as input information of the prompt text matching the object to be processed.

[0111] Furthermore, B3 may include:

[0112] B301. Determine, according to the defect flow information of the object to be processed, timing information of at least two defect processing processes of the object to be processed;

[0113] B302. Adding timing information to each defect processing process of the object to be processed;

[0114] B303. The defect flow information of the object to be processed after adding the timing information is used as input information of the prompt text matching the object to be processed.

[0115] Furthermore, B3 may include: extracting keywords from defect flow information of the object to be processed to obtain at least one keyword; and determining input information of a prompt text matching the object to be processed based on the extracted keywords and defect description information of the object to be processed.

[0116] Furthermore, the prompt text matching the object to be processed also includes instruction information; the instruction information is used to instruct the defect cause classification model to determine the defect cause type of the object to be processed from various defect cause types of the background information according to the input information.

[0117] Similarly, the process of generating background information of a prompt text matching the object to be processed based on the defect cause type, generating input information of a prompt text matching the object to be processed based on the defect description information of the object to be processed and the defect flow information, and generating instruction information of a prompt text matching the object to be processed is the same as the process of generating background information, input information and instruction information of the prompt text of the current sample object, and will not be repeated here in this embodiment.

[0118] S230: Input the prompt text matching the object to be processed into a defect cause classification model to obtain the defect cause type of the object to be processed output by the defect cause classification model.

[0119] The defect cause classification model is trained using the method of S110-S130.

[0120] In this embodiment, when constructing the prompt text as the input of the defect cause classification model, the defect description information, defect flow information, and the pre-set defect cause classification in the defect report of the defect management tool are combined, and the defect handling subject role information and the defect handling process timing information are added to the defect flow information, thereby realizing multi-dimensional data fusion and constructing a comprehensive defect data representation.

[0121] In an embodiment of the present invention, by obtaining a defect report of the object to be processed, a prompt text is determined according to the defect report and a preset defect cause type, and the prompt text is input into a defect cause classification model obtained by training a large model to obtain a defect cause type output by the defect cause classification model. This solves the problem in the prior art that when defect management tools are used to track defects, when defect states are transferred at different defect handling entities, the defect cause location and processing comments of different defect handling entities are inconsistent, resulting in unclear defect cause analysis results. This avoids the obstruction of human interference factors in tracking defect causes, and improves the accuracy and objectivity of defect cause analysis.

[0122] Figure 3 A schematic diagram of a structure of a device for determining a defect cause provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, the device for determining the cause of the defect includes:

[0123] A sample defect report determination module 310, for determining a sample defect report that matches at least two sample objects;

[0124] The current sample object prompt text determination module 320 is used to determine the prompt text matching the current sample object for each sample object according to at least two preset defect cause types and the sample defect report matching the current sample object;

[0125] A current sample data pair determination module 330 is used to construct a current sample data pair according to a prompt text matched with the current sample object and a defect cause type of the current sample object;

[0126] The defect cause classification model determination module 340 is used to train the large language model according to the sample data pairs corresponding to each sample object to obtain a defect cause classification model, so as to determine the defect cause according to the defect cause classification model.

[0127] In the embodiment of the present invention, by determining the sample defect report of each sample object, a prompt text is generated for the current sample object according to the preset defect cause type and the sample defect report of the current sample object, a current sample data pair is constructed according to the prompt text of the current sample object and its defect cause type, and a large language model is trained according to the sample data pairs corresponding to each sample object to obtain a defect cause classification model, so as to determine the defect cause according to the defect cause classification model. The problem of inconsistent defect cause positioning when defect tracking is performed through defect management tools in the prior art is solved, the defect cause classification model is trained based on the large model, the actual cause of the defect is comprehensively judged, the accuracy and rationality of defect cause classification are improved, and the obstacles to tracking the actual defect cause caused by human interference factors are reduced.

[0128] Based on the above embodiment, optionally, the current sample object prompt text determination module 320 includes:

[0129] A current sample object prompt text background information determination unit, used to use at least two preset defect cause types as background information of the prompt text matching the current sample object;

[0130] A current sample object defect information determination unit, configured to determine defect description information and defect flow information of the current sample object according to a sample defect report matching the current sample object;

[0131] The defect flow information is a record of the state changes of the defect of the current sample object in the entire life cycle and at least one defect handling process;

[0132] The current sample object prompt text input information determining unit is used to use the defect description information and defect flow information of the current sample object as input information of the prompt text matching the current sample object.

[0133] Based on the above embodiment, optionally, the current sample object prompt text input information determining unit is specifically used to:

[0134] Determine at least two defect handling subject identifiers in the defect flow information of the current sample object;

[0135] Add the role information of the defect handling subject to each defect handling subject identification;

[0136] The defect flow information of the current sample object after adding the defect processing subject role information is used as input information of the prompt text matching the current sample object.

[0137] Based on the above embodiment, optionally, the current sample object prompt text input information determining unit is specifically used to:

[0138] Determine, according to defect flow information of the current sample object, timing information of at least two defect handling processes of the current sample object;

[0139] Adding timing information to each defect processing process of the current sample object;

[0140] The defect flow information of the current sample object after adding the timing information is used as input information of the prompt text matching the current sample object.

[0141] Based on the above embodiment, optionally, the current sample object prompt text input information determining unit is specifically used to:

[0142] Extract keywords from defect flow information of the current sample object to obtain at least one keyword;

[0143] According to the extracted keywords and the defect description information of the current sample object, the input information of the prompt text matching the current sample object is determined.

[0144] Based on the above embodiment, optionally, the prompt text matching the current sample object further includes instruction information;

[0145] The instruction information is used to instruct the large language model to determine the defect cause type of the current sample object from among the defect cause types of the background information according to the input information.

[0146] The device for determining the cause of a defect provided in the embodiment of the present invention can execute the method for determining the cause of a defect provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0147] Figure 4 A schematic diagram of a structure of a device for determining a defect cause provided by an embodiment of the present invention is shown in FIG. Figure 4 As shown, the device for determining the cause of the defect includes:

[0148] A defect report determination module 410 for an object to be processed, for determining a defect report matching the object to be processed;

[0149] The prompt text determination module 420 for the object to be processed is used to determine the prompt text matching the object to be processed according to at least two preset defect cause types and the defect report matching the object to be processed;

[0150] The defect cause type determination module 430 of the object to be processed is used to input the prompt text matching the object to be processed into the defect cause classification model to obtain the defect cause type of the object to be processed output by the defect cause classification model;

[0151] The defect cause classification model is trained using the method in the above embodiment.

[0152] In an embodiment of the present invention, by obtaining a defect report of the object to be processed, a prompt text is determined according to the defect report and a preset defect cause type, and the prompt text is input into a defect cause classification model obtained by training a large model to obtain a defect cause type output by the defect cause classification model. This solves the problem in the prior art that when defect management tools are used to track defects, when defect states are transferred at different defect handling entities, the defect cause location and processing comments of different defect handling entities are inconsistent, resulting in unclear defect cause analysis results. This avoids the obstruction of human interference factors in tracking defect causes, and improves the accuracy and objectivity of defect cause analysis.

[0153] Based on the above embodiment, optionally, the prompt text determination module 420 of the object to be processed includes:

[0154] A background information determination unit for the prompt text of the object to be processed, used to use at least two preset defect cause types as background information of the prompt text matching the object to be processed;

[0155] A defect information determination unit for an object to be processed, configured to determine defect description information and defect flow information of the object to be processed according to the defect report matching the object to be processed;

[0156] The defect flow information is a record of the state changes of the defects of the object to be processed during the entire life cycle and at least one defect processing process;

[0157] The prompt text input information determining unit of the object to be processed is used to use the defect description information and the defect flow information of the object to be processed as the input information of the prompt text matching the object to be processed.

[0158] On the basis of the above embodiment, optionally, the prompt text input information determining unit of the object to be processed is specifically used for:

[0159] Determine at least two defect handling subject identifiers in the defect transfer information of the object to be handled;

[0160] Add the role information of the defect handling subject to each defect handling subject identification;

[0161] The defect flow information of the object to be processed after adding the defect processing subject role information is used as input information of the prompt text matching the object to be processed.

[0162] On the basis of the above embodiment, optionally, the prompt text input information determining unit of the object to be processed is specifically used for:

[0163] Determine, according to the defect flow information of the object to be processed, timing information of at least two defect processing processes of the object to be processed;

[0164] Adding timing information to each defect processing process of the object to be processed;

[0165] The defect flow information of the object to be processed after adding the timing information is used as input information of the prompt text matching the object to be processed.

[0166] On the basis of the above embodiment, optionally, the prompt text input information determining unit of the object to be processed is specifically used for:

[0167] Extract keywords from defect flow information of the object to be processed to obtain at least one keyword;

[0168] According to the extracted keywords and the defect description information of the object to be processed, the input information of the prompt text matching the object to be processed is determined.

[0169] Based on the above embodiment, optionally, the prompt text matching the object to be processed also includes instruction information;

[0170] The instruction information is used to instruct the defect cause classification model to determine the defect cause type of the object to be processed from among the defect cause types of the background information according to the input information.

[0171] The device for determining the cause of a defect provided in the embodiment of the present invention can execute the method for determining the cause of a defect provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0172] Figure 5 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0173] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0174] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0175] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as a workflow management method.

[0176] In some embodiments, the workflow management method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the workflow management method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the workflow management method in any other appropriate manner (e.g., by means of firmware).

[0177] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0178] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0179] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0180] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0181] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0182] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0183] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0184] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements a method as provided in any embodiment of the present application. . .

[0185] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0186] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for determining the cause of a defect, characterized in that: include: identifying sample defect reports that match at least two sample objects; For each sample object, according to at least two preset defect cause types and a sample defect report matching the current sample object, determine a prompt text matching the current sample object; Constructing a current sample data pair according to the prompt text matching the current sample object and the defect cause type of the current sample object; The large language model is trained according to the sample data pairs corresponding to each sample object to obtain a defect cause classification model, so as to determine the defect cause according to the defect cause classification model.

2. The method according to claim 1, characterized in that According to at least two preset defect cause types and a sample defect report matching the current sample object, a prompt text matching the current sample object is determined, including: At least two pre-set defect cause types are used as background information of the prompt text matching the current sample object; Determine defect description information and defect flow information of the current sample object according to the sample defect report matching the current sample object; The defect flow information is a record of the state changes of the defect of the current sample object in the entire life cycle and at least one defect handling process; The defect description information and defect flow information of the current sample object are used as input information of the prompt text matching the current sample object.

3. The method according to claim 2, characterized in that The defect flow information of the current sample object is used as input information of the prompt text matching the current sample object, including: Determine at least two defect handling subject identifiers in the defect flow information of the current sample object; Add the role information of the defect handling subject to each defect handling subject identification; The defect flow information of the current sample object after adding the defect processing subject role information is used as input information of the prompt text matching the current sample object.

4. The method according to claim 2, characterized in that: The defect flow information of the current sample object is used as input information of the prompt text matching the current sample object, including: Determine, according to defect flow information of the current sample object, timing information of at least two defect handling processes of the current sample object; Adding timing information to each defect processing process of the current sample object; The defect flow information of the current sample object after adding the timing information is used as input information of the prompt text matching the current sample object.

5. The method according to claim 2, characterized in that: The defect description information and defect flow information of the current sample object are used as input information of the prompt text matching the current sample object, including: Extract keywords from defect flow information of the current sample object to obtain at least one keyword; According to the extracted keywords and the defect description information of the current sample object, the input information of the prompt text matching the current sample object is determined.

6. The method according to claim 1, characterized in that The prompt text matching the current sample object also includes instruction information; The instruction information is used to instruct the large language model to determine the defect cause type of the current sample object from among the defect cause types of the background information according to the input information.

7. A method for determining the cause of a defect, characterized in that: The method comprises: Determine the defect report that matches the object to be processed; Determine a prompt text matching the object to be processed according to at least two preset defect cause types and the defect report matching the object to be processed; Inputting the prompt text matching the object to be processed into a defect cause classification model to obtain the defect cause type of the object to be processed output by the defect cause classification model; Wherein, the defect cause classification model is trained using the method described in any one of claims 1-6.

8. The method according to claim 7, characterized in that According to at least two preset defect cause types and the defect report matching the object to be processed, a prompt text matching the object to be processed is determined, including: At least two pre-set defect cause types are used as background information of the prompt text matching the object to be processed; Determine defect description information and defect flow information of the object to be processed according to the defect report matching the object to be processed; The defect flow information is a record of the state changes of the defects of the object to be processed during the entire life cycle and at least one defect processing process; The defect description information and the defect flow information of the object to be processed are used as input information of the prompt text matching the object to be processed.

9. The method according to claim 8, characterized in that The defect flow information of the object to be processed is used as input information of a prompt text matching the object to be processed, including: Determine at least two defect handling subject identifiers in the defect transfer information of the object to be handled; Add the role information of the defect handling subject to each defect handling subject identification; The defect flow information of the object to be processed after adding the defect processing subject role information is used as input information of the prompt text matching the object to be processed.

10. The method according to claim 8, characterized in that The defect flow information of the object to be processed is used as input information of a prompt text matching the object to be processed, including: Determine, according to the defect flow information of the object to be processed, timing information of at least two defect processing processes of the object to be processed; Adding timing information to each defect processing process of the object to be processed; The defect flow information of the object to be processed after adding the timing information is used as input information of the prompt text matching the object to be processed.

11. The method according to claim 8, characterized in that The defect description information and defect flow information of the object to be processed are used as input information of a prompt text matching the object to be processed, including: Extract keywords from defect flow information of the object to be processed to obtain at least one keyword; According to the extracted keywords and the defect description information of the object to be processed, the input information of the prompt text matching the object to be processed is determined.

12. The method according to claim 7, characterized in that The prompt text matching the object to be processed also includes instruction information; The instruction information is used to instruct the defect cause classification model to determine the defect cause type of the object to be processed from among the defect cause types of the background information according to the input information.

13. A device for determining the cause of a defect, characterized in that: include: A sample defect report determination module, used to determine a sample defect report that matches at least two sample objects; A current sample object prompt text determination module is used to determine, for each sample object, a prompt text matching the current sample object according to at least two preset defect cause types and a sample defect report matching the current sample object; A current sample data pair determination module, used to construct a current sample data pair according to a prompt text matched with a current sample object and a defect cause type of the current sample object; The defect cause classification model determination module is used to train the large language model according to the sample data pairs corresponding to each sample object to obtain the defect cause classification model, so as to determine the defect cause according to the defect cause classification model.

14. A device for determining the cause of a defect, characterized in that: include: A defect report determination module for an object to be processed, used to determine a defect report matching the object to be processed; A prompt text determination module for an object to be processed, used to determine a prompt text matching the object to be processed according to at least two preset defect cause types and the defect report matching the object to be processed; A module for determining the defect cause type of the object to be processed, used for inputting the prompt text matching the object to be processed into the defect cause classification model to obtain the defect cause type of the object to be processed output by the defect cause classification model; Wherein, the defect cause classification model is trained using the method described in any one of claims 1-6.

15. An electronic device for determining the cause of a defect, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method for determining the cause of a defect as described in any one of claims 1 to 6, or implements the method for determining the cause of a defect as described in any one of claims 7 to 12.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, it implements the method for determining the cause of a defect as described in any one of claims 1 to 6, or it implements the method for determining the cause of a defect as described in any one of claims 7 to 12.

17. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the method for determining the cause of a defect as described in any one of claims 1 to 6, or implements the method for determining the cause of a defect as described in any one of claims 7 to 12.

Images