A software release risk prediction method, device, equipment and medium

Abstract

The present application relates to the technical field of software risk, and discloses a software release risk prediction method, device, equipment and medium, comprising: obtaining a plurality of inventory projects; based on a plurality of inventory project problem labels, obtaining a plurality of risk characteristics corresponding to a plurality of inventory risk problems; mapping the plurality of risk characteristics into a plurality of vector characteristics and storing them in a database; obtaining a project to be reviewed, establishing a plurality of similar vector characteristics from the database through clustering analysis; performing semantic processing to obtain a plurality of prompt sentences; and responding the plurality of prompt sentences to a review page. Through the method of mathematical analysis, the possible risks of the new version project can be efficiently estimated, and the review page can be directly responded, avoiding the omissions caused by relying on manual maintenance.

Description

Technical Field

[0001] The present invention relates to the field of software risk technology, and in particular to a software release risk prediction method, device, equipment and medium. Background Art

[0002] In the era of rapid mobile internet development, the complexity of software projects is rapidly increasing. Security, compliance, and high-quality delivery have become essential capabilities for the development of internet companies and software-based high-tech companies. Therefore, improving the quality of production releases during software iteration has become a key focus in software development.

[0003] In today's software development field, the production release process often improves release quality by maintaining a pre-release checklist and a release process checklist. However, manual maintenance and review can lead to increased costs, blind spots, and various unexpected omissions, thus posing a production release quality risk. Summary of the Invention

[0004] The present invention provides a software release risk prediction method, device, equipment and medium to solve the technical problem that manual maintenance and inspection of software to be released are prone to various omissions, thereby constituting production and release quality risks.

[0005] In a first aspect, a software release risk prediction method is provided, comprising:

[0006] Acquire a plurality of stock items, and extract a plurality of stock version numbers, a plurality of stock item metadata, a plurality of stock item issues, and a plurality of stock item issue tags corresponding to the plurality of stock item issues from the plurality of stock items;

[0007] Based on the plurality of stock project problem labels, establishing a plurality of stock risk problems from the plurality of stock project problems to obtain a plurality of risk features corresponding to the plurality of stock risk problems;

[0008] Mapping the risk features into vector features, and storing the vector features, the stock version numbers, and the stock item metadata into a database;

[0009] Obtaining a project to be reviewed, extracting a version number to be reviewed and meta information of the project to be reviewed from the project to be reviewed, and establishing a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed;

[0010] establishing the risk features corresponding to the similar vector features as similar risk features, and performing semantic processing on the similar risk features to obtain prompt statements;

[0011] A plurality of the prompt statements are responded to the review page to evaluate the software release risk of the project to be reviewed.

[0012] In a second aspect, a software release risk prediction device is provided, comprising:

[0013] an acquisition module, configured to acquire a plurality of stock items, and extract a plurality of stock version numbers, a plurality of stock item metadata, a plurality of stock item problems, and a plurality of stock item problem tags corresponding to the plurality of stock item problems from the plurality of stock items;

[0014] a selection module configured to identify a plurality of stock risk issues from the plurality of stock project issues based on the plurality of stock project issue labels, so as to obtain a plurality of risk features corresponding to the plurality of stock risk issues;

[0015] a mapping module, configured to map the risk features into vector features, and store the vector features, the stock version numbers, and the stock item metadata into a database;

[0016] an analysis module, configured to obtain a project to be reviewed, extract a version number to be reviewed and meta information of the project to be reviewed from the project to be reviewed, and establish a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed;

[0017] a semantic module, configured to establish the risk features corresponding to the similar vector features as similar risk features, and perform semantic processing on the similar risk features to obtain prompt statements;

[0018] The response module is used to respond to the plurality of prompt statements to the review page to evaluate the software release risk of the project to be reviewed.

[0019] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned software release risk prediction method when executing the computer program.

[0020] In a fourth aspect, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned software release risk prediction method are implemented.

[0021] In the scheme implemented by the above-mentioned software release risk prediction method, device, computer equipment and storage medium, several historical version projects, i.e., several stock projects, can be obtained, and various types of project-related information can be obtained, including online problems that have occurred in the project, and stock project problems. The stock project problems are associated with stock project problem labels, which can be used to identify risky problems through the stock project problem labels; several risk features in several stock projects are extracted, and their risks are evaluated by numerical quantification to form several vector features, which are convenient for analysis by mathematical methods; several vector features are stored and related information, so that when evaluating a new version of the software, similar projects can be retrieved by querying the database and obtaining several similar vector features; by semanticizing several similar risk features, the risk assessment of the production and release of the new version of the software can be intuitively displayed in the form of statements. In the present invention, project problems and related data are recorded during the software iteration process, and risk features are extracted and quantified. Through mathematical analysis methods, the possible risks of the new version project can be efficiently estimated and intuitively responded to on the review page, avoiding the problem of relying solely on manual maintenance that is prone to omissions and thus causing release quality risks, and reducing labor costs and inspection blind spots. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0023] Figure 1 This is a schematic diagram of an application environment of a software release risk prediction method according to an embodiment of the present invention;

[0024] Figure 2 This is a flow chart of a software release risk prediction method according to an embodiment of the present invention;

[0025] Figure 3 This is a structural diagram of a software release risk prediction device in one embodiment of the present invention;

[0026] Figure 4 is a structural diagram of a computer device in one embodiment of the present invention;

[0027] Figure 5 FIG. 2 is another structural diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0029] The software release risk prediction method provided by the embodiment of the present invention can be applied in the following situations: Figure 1 In an application environment, the client communicates with the server through a network. The server can obtain a number of stock items through the client, extract a number of stock version numbers, a number of stock item metadata, a number of stock item problems and a number of stock item problem labels corresponding to the number of stock item problems from the stock items; based on the number of stock item problem labels, establish a number of stock risk issues from the stock item problems to obtain a number of risk features corresponding to the number of stock risk issues; map the number of risk features into a number of vector features, and store the number of vector features, the stock version numbers and the stock item metadata into a database; obtain the project to be reviewed, extract the version number to be reviewed and the metadata of the project to be reviewed from the project to be reviewed, and based on the version number to be reviewed and the The metadata of the project to be reviewed is used to establish several similar vector features from the database through cluster analysis; several risk features corresponding to the several similar vector features are established as several similar risk features, and several similar risk features are semantically processed to obtain several prompt statements; several prompt statements are responded to the review page to evaluate the software release risk of the project to be reviewed. In the present invention, project problems and related data are recorded during the software iteration process, and risk features are extracted and quantified. Through mathematical analysis methods, the possible risks of the new version of the project can be efficiently estimated and intuitively responded to the review page, avoiding the problem of relying solely on manual maintenance that is prone to omissions and thus causing release quality risks, and reducing labor costs and inspection blind spots. Among them, the client can be but is not limited to various personal computers, laptops, smart phones, tablets and portable wearable devices. The server can be implemented with an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments.

[0030] See also Figure 2 As shown, Figure 2 A flowchart of a software release risk prediction method provided by an embodiment of the present invention includes the following steps:

[0031] S10: Acquire a number of stock items, and extract a number of stock version numbers, a number of stock item metadata, a number of stock item problems, and a number of stock item problem tags corresponding to the number of stock item problems from the stock items.

[0032] The software release risk prediction method provided by the present invention can be applied to the iterative process of enterprise software development. The acquisition of several of the stock items can be achieved through the server, and the server can obtain historical version items of software production, that is, stock items. Each of the historical version items has a corresponding unique version number, that is, each of the stock items has a corresponding unique stock version number.

[0033] It should be noted that the existing project problems include online problems of released historical version projects, and may also include defect problems found in historical version test projects that have not been released for production. The existing project problem labels include whether it is a network problem, whether it is a disk problem, whether it is a business third-party problem, whether it is a configuration file problem, whether it is an R&D specification problem, etc. The existing project problem labels can also include problem labels customized by R&D personnel.

[0034] In S10, the stock item meta-information includes the problem record primary key, stock item name, problem cause, problem solution, improvement plan, and occurrence time.

[0035] After being extracted by the server, the plurality of stock version numbers, the plurality of stock item metadata, the plurality of stock item questions and the plurality of stock item question tags corresponding to the plurality of stock item questions form a data set and are stored.

[0036] S20: Based on the plurality of existing project problem labels, a plurality of existing risk problems are established from the plurality of existing project problems to obtain a plurality of risk features corresponding to the plurality of existing risk problems.

[0037] The content marked in the existing project problem tag is related to the occurrence factors and influencing factors of the existing project problem. It can be understood that by examining all the existing project problem tags, multiple risk categories can be extracted and summarized.

[0038] It should be noted that the basis for screening risks is whether they affect the normal delivery of product functional value, and test defects identified through the software testing process during software iteration can be clearly identified as risks. In S20, based on the multiple existing project problem labels, multiple existing risk problems are established from the multiple existing project problems to obtain multiple risk features corresponding to the multiple existing risk problems. Specifically, the following steps S21-S23 are included:

[0039] S21: Constructing a plurality of risk categories based on the plurality of problem labels of the existing projects.

[0040] In this implementation, the risk categories can be divided into technical risk, quality risk, cost risk, time risk, legal and compliance risk, and user acceptance risk.

[0041] S22: Establishing several of the existing project issues that meet the several risk categories as several of the existing risk issues.

[0042] In the embodiment of the present invention, some of the existing project problems are not caused by the software itself, such as third-party problems and device disk problems. The first level screening of the existing project problems is performed using the risk categories.

[0043] S23: Extracting a plurality of influencing factors from the plurality of existing risk issues, wherein the influencing factors constitute risk characteristics.

[0044] In an embodiment of the present invention, a second-level screening is performed on a number of the existing risk issues to specifically examine the possible influencing factors that may exist when the risk category causes the software risk, so as to extract and establish the risk characteristics.

[0045] Taking technical risk as an example, the risk characteristics associated with the technical risk category include technical difficulty, technical obsolescence and compatibility issues. Among them, technical difficulty means that some technologies may be very new or complex, resulting in greater difficulty in implementation. Technical obsolescence means that the selected technology may become obsolete during the project cycle, affecting the long-term maintenance and support of the software. Compatibility issues mean that compatibility issues between different software and hardware may lead to incomplete functions or degraded performance.

[0046] Taking quality risk as an example, the risk characteristics associated with the quality risk category include functional defects, performance deficiencies and security vulnerabilities. Among them, functional defects mean that the software may have undiscovered functional errors or defects, affecting the user experience. Performance deficiencies mean that the software performs poorly under high load, such as long response time and high resource consumption. Security vulnerabilities mean that the software may have security vulnerabilities and is vulnerable to attacks.

[0047] Taking cost risk as an example, the risk characteristics associated with the cost risk category include budget overruns and low return on investment. Budget overruns mean that actual development costs exceed expectations, including labor costs, material costs, etc. Low return on investment means that the software may not achieve the expected business value or return on investment after completion.

[0048] Taking time risk as an example, the risk characteristics associated with the time risk category include delayed delivery and missed market windows. Delayed delivery means that the project may be delayed due to various reasons (such as changes in requirements and technical difficulties), and missed market windows means that if the software cannot be completed on time, the best time to enter the market may be missed.

[0049] Taking legal and compliance risks as an example, the risk characteristics associated with the legal and compliance risk category include copyright infringement, privacy protection, and non-compliance with standards. Copyright infringement means that the use of unauthorized code or resources may lead to legal disputes, privacy protection means that the failure to properly handle user data may violate relevant laws and regulations, and non-compliance with standards means that the software may not meet industry standards or regulatory requirements.

[0050] Taking user acceptance risk as an example, the risk characteristics associated with the user acceptance risk category include poor user experience, high training costs, and poor adaptability. Among them, poor user experience means that unreasonable design or unfriendly interface may lead to user dissatisfaction, high training costs mean that if the software operation is complex, higher user training costs may be required, and poor adaptability means that the software may not be suitable for the needs or workflow of the target user group.

[0051] Taking organizational management risk as an example, the risk characteristics associated with the organizational management risk category include poor team communication, leadership decision-making errors, and improper resource allocation. Among them, poor team communication means that poor communication between project team members may lead to misunderstandings and conflicts. Leadership decision-making errors mean that management decision-making errors, such as incorrect assessment of market demand, may lead to project failure. Improper resource allocation means that key resources (such as talent and funds) are irrationally allocated, affecting project progress.

[0052] The above embodiment can reflect that the software release risk prediction method can accurately screen out comprehensive and general risks by examining a large number of existing project problem tags, conducting more detailed classification and step-by-step screening, and improving the identification and screening of risks during the software iteration process, which is beneficial to making accurate risk assessment judgments when reviewing new projects, and avoids manual maintenance, effectively preventing inspection blind spots and omissions, and saving a lot of labor costs.

[0053] S30: Mapping the risk features into vector features, and storing the vector features, the existing version numbers, and the metadata of the existing items into a database;

[0054] Each existing item corresponds to a vector feature, and a numerical risk score is assigned to the existing item using the vector feature. The obtained vector features and related information about the existing items are promptly stored, making it easier for the server to retrieve the information needed to review new items through query operations.

[0055] In S30, the risk features are mapped into vector features, including the following steps S31-S32:

[0056] S31: Constructing a scoring system based on the plurality of risk features, so that each risk feature is associated with a scoring value.

[0057] In this embodiment, a scoring system is defined for each risk feature, for example, using a scale of 1 to 5, where 1 represents very low risk and 5 represents very high risk.

[0058] S32: Using the plurality of rating values ​​as elements of a vector to construct a plurality of vector features for the plurality of stock items.

[0059] A vector feature is created for each existing project, with each element of the vector corresponding to a risk feature score. Assume that the risk features of a project are technical difficulty, functional defects, budget overruns, delayed delivery, copyright infringement, poor user experience, and poor team communication. Technical difficulty is scored as 3, functional defects as 2, budget overruns as 4, delayed delivery as 3, copyright infringement as 1, poor user experience as 2, and poor team communication as 3. The constructed vector feature is [3, 2, 4, 3, 1, 2, 3].

[0060] S40: Acquire the project to be reviewed, extract the version number to be reviewed and the meta information of the project to be reviewed from the project to be reviewed, and establish a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed.

[0061] In this embodiment, the data type in the metadata of the project to be reviewed is consistent with the metadata of the existing project, and the version number to be reviewed is used to search for the most recent historical version number in the database.

[0062] Before S40, i.e., obtaining the project to be reviewed, extracting the version number to be reviewed and the meta information of the project to be reviewed from the project to be reviewed, and establishing a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed, the following steps are included:

[0063] S41: Calculate the distance between the vector features of two different stock items to establish the risk similarity between the vector features of the two different stock items.

[0064] Various mathematical tools can be used to analyze the vector features. For example, in the distance calculation in this embodiment, based on the database, the Euclidean distance or Manhattan distance between different items is calculated pairwise, which can be used to evaluate their risk similarity.

[0065] In S40, based on the version number to be reviewed and the meta information of the project to be reviewed, a plurality of similar vector features are determined from the database through cluster analysis, including steps S42-S47:

[0066] S42: Based on the version number to be reviewed, determine whether there is a similar existing version number in the database.

[0067] S43: If it exists, the vector feature corresponding to the stock version number is selected and established as the target vector feature, and similarity matching is performed based on the risk similarity and clustering algorithm to divide all the stock items in the database and obtain a group of items with similar risks to the target vector feature.

[0068] S44: If not, based on the metadata of the project to be reviewed, determine whether there is an existing project of the same category as the project to be reviewed in the database.

[0069] S45: If so, the stock project of the same category as the project to be reviewed will be established as the target project, and the vector feature corresponding to the target project will be established as the target vector feature. Similarity matching will be performed based on the risk similarity and clustering algorithm to divide all the stock projects in the database and obtain a project group with similar risks to the target vector feature.

[0070] S46: If it does not exist, then extract the test risk features from the test risk issues of the test defects generated during the development process of the project to be reviewed, map the test risk features into test vector features, and establish the test vector features as target vector features. Perform similarity matching based on the risk similarity and clustering algorithm to divide all the stock projects in the database and obtain a project group with similar risks to the target vector features.

[0071] S47: establishing the plurality of vector features corresponding to the item group as a plurality of similar vector features.

[0072] In steps S42-S47 of the embodiment of the present invention, after obtaining the project to be reviewed, the project most closely related to the project to be reviewed is searched, and priority is given to examining whether there is a historical version project of the project to be reviewed. If so, the historical version project closest to the version of the project to be reviewed is selected, and multiple similar vector features are obtained through its vector features. If there is no historical version project, the same category project is searched. For example, if the project to be reviewed obtained by a certain enterprise is a back-end project, the back-end project developed by the same enterprise is searched. If there is no historical version project and no project of the same type, the vector features are obtained through the test version closest to the release version during the development process of the project to be reviewed to obtain the target vector features.

[0073] A clustering algorithm, such as the K-means algorithm, is used to group the projects in the database and find vector features similar to the target vector features. The machine learning model can be trained to predict the risk level of the project to be reviewed, and several existing projects corresponding to several similar vector features, i.e., the project group, can be found. This helps to further extract solutions to the problems based on the existing metadata, and then recommend solutions to prevent risks.

[0074] S50: establishing the risk features corresponding to the similar vector features as similar risk features, and performing semantic processing on the similar risk features to obtain prompt statements.

[0075] In this embodiment, the prompt sentence is for example: This project has recently had coding problems, please supplement and review the relevant use cases. Using a semantic approach, the evaluation of the project to be reviewed will be more intuitive and clear.

[0076] S60: Sending a plurality of prompt statements to the review page to evaluate the software release risk of the project to be reviewed.

[0077] In this embodiment, the statements associated with the three risk features with the highest scores are preferentially selected for response and display. After S60, that is, after responding to the review page with several prompt statements to assess the software release risk of the project to be reviewed, steps S61-S62 are also included:

[0078] S61: Based on the plurality of similar vector features, establish a plurality of similar item meta-information from the plurality of stock item meta-information in the database.

[0079] S62: extracting recommended prevention solutions from the meta-information of a plurality of similar projects, and responding the recommended prevention solutions to the review page.

[0080] In this embodiment, steps S61-S62 further search for the metadata of the historical project associated with the similar vector feature through the similar vector feature, that is, the similar project metadata, and extract the cause of the problem, the solution to the problem and the improvement plan from the similar project metadata, thereby obtaining the recommended prevention plan, which corresponds one-to-one with the prompt statement and is displayed together on the review page.

[0081] It can be seen that in the above solution, project problems and related data are recorded during the software iteration process, and risk characteristics are extracted and quantified, so that complex software risk management problems are converted into mathematical problems. Current data analysis and machine learning technologies can be used for more effective management, and the possible risks of new version projects can be efficiently estimated and responded to intuitively on the review page, avoiding the problem of relying solely on manual maintenance that is prone to omissions and thus causing release quality risks, and reducing labor costs and inspection blind spots.

[0082] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0083] In one embodiment, a software release risk prediction device is provided, which corresponds to the software release risk prediction method in the above embodiment. Figure 3 As shown, the software release risk prediction device includes an acquisition module 101, a selection module 102, a mapping module 103, an analysis module 104, a semantic module 105 and a response module 106. The functional modules are described in detail as follows:

[0084] An acquisition module 101 is configured to acquire a plurality of stock items and extract a plurality of stock version numbers, a plurality of stock item metadata, a plurality of stock item issues, and a plurality of stock item issue tags corresponding to the plurality of stock item issues from the plurality of stock items;

[0085] A selection module 102 is configured to identify a plurality of existing risk issues from the plurality of existing project issues based on the plurality of existing project issue labels, so as to obtain a plurality of risk features corresponding to the plurality of existing risk issues;

[0086] A mapping module 103 is configured to map the risk features into vector features, and store the vector features, the stock version numbers, and the stock item metadata into a database;

[0087] An analysis module 104 is configured to obtain a project to be reviewed, extract a version number to be reviewed and meta information of the project to be reviewed from the project to be reviewed, and establish a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed;

[0088] The semantic module 105 is configured to establish the risk features corresponding to the similar vector features as similar risk features, and perform semantic processing on the similar risk features to obtain prompt statements.

[0089] The response module 106 is configured to respond to the plurality of prompt statements on the review page to assess the software release risk of the project to be reviewed.

[0090] In one embodiment, the acquisition module 101 is specifically configured to:

[0091] The stock item meta-information includes the problem record primary key, stock item name, problem cause, problem solution, improvement plan, and occurrence time.

[0092] In one embodiment, the selection module 102 is specifically configured to:

[0093] Constructing a plurality of risk categories based on a plurality of said stock project problem labels;

[0094] Establishing a number of said stock project issues that meet said number of said risk categories as a number of said stock risk issues;

[0095] A number of influencing factors are extracted from the number of existing risk issues, and the influencing factors constitute risk characteristics.

[0096] In one embodiment, the mapping module 103 is specifically configured to:

[0097] Constructing a scoring system based on the plurality of risk characteristics so that each risk characteristic is associated with a scoring value;

[0098] A plurality of the rating values ​​are used as elements of a vector to construct a plurality of the vector features for a plurality of the inventory items.

[0099] In one embodiment, the analysis module 104 is specifically configured to:

[0100] The distance between the vector features of two different stock items is calculated to establish the risk similarity between the vector features of the two different stock items.

[0101] In one embodiment, the analysis module 104 is further configured to:

[0102] Based on the version number to be reviewed, determining whether there is a similar existing version number in the database;

[0103] If so, the vector feature corresponding to the existing version number is selected and established as the target vector feature. Similarity matching is performed based on the risk similarity and clustering algorithm to divide all the existing projects in the database and obtain a group of projects with similar risks to the target vector feature.

[0104] If not, judging whether there is the stock project of the same category as the project to be reviewed in the database based on the meta information of the project to be reviewed;

[0105] If so, the existing project of the same category as the project to be reviewed is established as the target project, and the vector feature corresponding to the target project is established as the target vector feature. Similarity matching is performed based on the risk similarity and clustering algorithm to divide all the existing projects in the database and obtain a project group with similar risks to the target vector feature;

[0106] If not, extracting test risk features from the test risk issues of test defects generated during the development process of the project to be reviewed, mapping the test risk features into test vector features, and establishing the test vector features as target vector features, performing similarity matching based on the risk similarity and clustering algorithm to divide all the stock projects in the database and obtain a project group with similar risks to the target vector features;

[0107] A plurality of the vector features corresponding to the item group are established as a plurality of similar vector features.

[0108] In one embodiment, the response module 106 is specifically configured to:

[0109] Based on the plurality of similar vector features, establishing a plurality of similar item meta-information from the plurality of stock item meta-information in the database;

[0110] Extracting recommended prevention solutions from the meta-information of a plurality of similar projects, and responding the recommended prevention solutions to the review page.

[0111] The present invention provides a software release risk prediction device, which first obtains a number of existing projects and extracts relevant information, obtains a number of risk features based on the problem labels of the several existing projects, maps the several risk features into a number of vector features, and stores them together with the relevant information in a database, then obtains the project to be reviewed, establishes a number of similar vector features through cluster analysis, and establishes a number of similar risk features, performs semantic processing, and responds to the review page. It can effectively estimate the possible risks of the new version project through mathematical analysis methods and respond intuitively to the review page, avoiding the problem of relying solely on manual maintenance that is prone to omissions and thus causing release quality risks, and reducing labor costs and inspection blind spots.

[0112] The specific limitations of the software release risk prediction device can be found in the limitations of the software release risk prediction method described above and will not be repeated here. Each module in the aforementioned software release risk prediction device can be implemented in whole or in part through software, hardware, or a combination thereof. Each of the aforementioned modules can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in a computer device in software form, so that the processor can call and execute the operations corresponding to each of the aforementioned modules.

[0113] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, memory, network interface and database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, it implements the functions or steps on the server side of a software release risk prediction method.

[0114] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 5As shown. The computer device includes a processor, memory, network interface, display screen, and input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When executed by the processor, the computer program implements the functions or steps on the client side of a software release risk prediction method.

[0115] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed:

[0116] Acquire a plurality of stock items, and extract a plurality of stock version numbers, a plurality of stock item metadata, a plurality of stock item issues, and a plurality of stock item issue tags corresponding to the plurality of stock item issues from the plurality of stock items;

[0117] Based on the plurality of stock project problem labels, establishing a plurality of stock risk problems from the plurality of stock project problems to obtain a plurality of risk features corresponding to the plurality of stock risk problems;

[0118] Mapping the risk features into vector features, and storing the vector features, the stock version numbers, and the stock item metadata into a database;

[0119] Obtaining a project to be reviewed, extracting a version number to be reviewed and meta information of the project to be reviewed from the project to be reviewed, and establishing a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed;

[0120] establishing the risk features corresponding to the similar vector features as similar risk features, and performing semantic processing on the similar risk features to obtain prompt statements;

[0121] A plurality of the prompt statements are responded to the review page to evaluate the software release risk of the project to be reviewed.

[0122] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0123] Acquire a plurality of stock items, and extract a plurality of stock version numbers, a plurality of stock item metadata, a plurality of stock item issues, and a plurality of stock item issue tags corresponding to the plurality of stock item issues from the plurality of stock items;

[0124] Based on the plurality of stock project problem labels, establishing a plurality of stock risk problems from the plurality of stock project problems to obtain a plurality of risk features corresponding to the plurality of stock risk problems;

[0125] Mapping the risk features into vector features, and storing the vector features, the stock version numbers, and the stock item metadata into a database;

[0126] Obtaining a project to be reviewed, extracting a version number to be reviewed and meta information of the project to be reviewed from the project to be reviewed, and establishing a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed;

[0127] establishing the risk features corresponding to the similar vector features as similar risk features, and performing semantic processing on the similar risk features to obtain prompt statements;

[0128] A plurality of the prompt statements are responded to the review page to evaluate the software release risk of the project to be reviewed.

[0129] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can be found in the relevant descriptions of the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0130] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0131] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0132] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A software release risk prediction method, characterized in that: include: Acquire a plurality of stock items, and extract a plurality of stock version numbers, a plurality of stock item metadata, a plurality of stock item issues, and a plurality of stock item issue tags corresponding to the plurality of stock item issues from the plurality of stock items; Based on the plurality of stock project problem labels, establishing a plurality of stock risk problems from the plurality of stock project problems to obtain a plurality of risk features corresponding to the plurality of stock risk problems; Mapping the risk features into vector features, and storing the vector features, the stock version numbers, and the stock item metadata into a database; Obtaining a project to be reviewed, extracting a version number to be reviewed and meta information of the project to be reviewed from the project to be reviewed, and establishing a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed; The step of establishing a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed includes: Based on the version number to be reviewed, determining whether there is a similar existing version number in the database; If the existing version number exists, the vector feature corresponding to the existing version number is selected and established as the target vector feature. Similarity matching is performed based on the risk similarity and clustering algorithm to divide all the existing projects in the database and obtain a group of projects with similar risks to the target vector feature. If the existing version number does not exist, determining whether there is an existing project of the same category as the project to be reviewed in the database based on the metadata of the project to be reviewed; If the existing project exists, the existing project of the same category as the project to be reviewed is established as a target project, and the vector feature corresponding to the target project is established as a target vector feature. Similarity matching is performed based on the risk similarity and clustering algorithm to divide all the existing projects in the database and obtain a project group with similar risks to the target vector feature; If the existing project does not exist, extracting test risk features from the test risk issues that generate test defects during the development process of the project to be reviewed, mapping the test risk features into test vector features, and establishing the test vector features as target vector features. Based on the risk similarity and clustering algorithm, similarity matching is performed to divide all the existing projects in the database, and obtain a project group with similar risks to the target vector features; establishing a plurality of the vector features corresponding to the project group as a plurality of similar vector features; establishing the risk features corresponding to the similar vector features as similar risk features, and performing semantic processing on the similar risk features to obtain prompt statements; A plurality of the prompt statements are responded to the review page to evaluate the software release risk of the project to be reviewed.

2. The software release risk prediction method according to claim 1, wherein: The stock item meta-information includes the problem record primary key, stock item name, problem cause, problem solution, improvement plan, and occurrence time.

3. The software release risk prediction method according to claim 1, wherein: The step of establishing a plurality of existing risk issues from the plurality of existing project issues based on the plurality of existing project issue labels to obtain a plurality of risk features corresponding to the plurality of existing risk issues includes: Constructing a plurality of risk categories based on a plurality of said stock project problem labels; Establishing a number of said stock project issues that meet said number of said risk categories as a number of said stock risk issues; A number of influencing factors are extracted from the number of existing risk issues, and the influencing factors constitute risk characteristics.

4. The software release risk prediction method according to claim 1, wherein: The step of mapping the plurality of risk features into a plurality of vector features includes: Constructing a scoring system based on the plurality of risk characteristics so that each risk characteristic is associated with a scoring value; A plurality of the rating values ​​are used as elements of a vector to construct a plurality of the vector features for a plurality of the inventory items.

5. The software release risk prediction method according to claim 1, wherein: The steps of obtaining the project to be reviewed, extracting the version number to be reviewed and the meta information of the project to be reviewed from the project to be reviewed, and establishing a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed may also include: The distance between the vector features of two different stock items is calculated to establish the risk similarity between the vector features of the two different stock items.

6. The software release risk prediction method according to claim 1, wherein: After the step of responding the plurality of prompt statements to the review page to assess the software release risk of the project to be reviewed, the following steps are further performed: Based on the plurality of similar vector features, establishing a plurality of similar item meta-information from the plurality of stock item meta-information in the database; Extracting recommended prevention solutions from the meta-information of a plurality of similar projects, and responding the recommended prevention solutions to the review page.

7. A software release risk prediction device, characterized in that: include: an acquisition module, configured to acquire a plurality of stock items, and extract a plurality of stock version numbers, a plurality of stock item metadata, a plurality of stock item problems, and a plurality of stock item problem tags corresponding to the plurality of stock item problems from the plurality of stock items; a selection module configured to identify a plurality of stock risk issues from the plurality of stock project issues based on the plurality of stock project issue labels, so as to obtain a plurality of risk features corresponding to the plurality of stock risk issues; a mapping module, configured to map the risk features into vector features, and store the vector features, the stock version numbers, and the stock item metadata into a database; an analysis module, configured to obtain a project to be reviewed, extract a version number to be reviewed and meta information of the project to be reviewed from the project to be reviewed, and establish a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed; The step of establishing a plurality of similar vector features from the database through cluster analysis based on the version number to be reviewed and the meta information of the project to be reviewed includes: Based on the version number to be reviewed, determining whether there is a similar existing version number in the database; If the existing version number exists, the vector feature corresponding to the existing version number is selected and established as the target vector feature. Similarity matching is performed based on the risk similarity and clustering algorithm to divide all the existing projects in the database and obtain a group of projects with similar risks to the target vector feature. If the existing version number does not exist, determining whether there is an existing project of the same category as the project to be reviewed in the database based on the metadata of the project to be reviewed; If the existing project exists, the existing project of the same category as the project to be reviewed is established as a target project, and the vector feature corresponding to the target project is established as a target vector feature. Similarity matching is performed based on the risk similarity and clustering algorithm to divide all the existing projects in the database and obtain a project group with similar risks to the target vector feature; If the existing project does not exist, extracting test risk features from the test risk issues that generate test defects during the development process of the project to be reviewed, mapping the test risk features into test vector features, and establishing the test vector features as target vector features. Based on the risk similarity and clustering algorithm, similarity matching is performed to divide all the existing projects in the database, and obtain a project group with similar risks to the target vector features; establishing a plurality of the vector features corresponding to the project group as a plurality of similar vector features; a semantic module, configured to establish the risk features corresponding to the similar vector features as similar risk features, and perform semantic processing on the similar risk features to obtain prompt statements; The response module is used to respond to the plurality of prompt statements to the review page to evaluate the software release risk of the project to be reviewed.

8. A computer 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 computer program, the steps of the software release risk prediction method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the software release risk prediction method according to any one of claims 1 to 6 are implemented.

Images