CN109299381B - Software defect retrieval and analysis system and method based on semantic concept - Google Patents
Software defect retrieval and analysis system and method based on semantic concept Download PDFInfo
- Publication number
- CN109299381B CN109299381B CN201811285850.5A CN201811285850A CN109299381B CN 109299381 B CN109299381 B CN 109299381B CN 201811285850 A CN201811285850 A CN 201811285850A CN 109299381 B CN109299381 B CN 109299381B
- Authority
- CN
- China
- Prior art keywords
- software
- user
- module
- defect
- label
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/40—Transformation of program code
- G06F8/41—Compilation
- G06F8/43—Checking; Contextual analysis
- G06F8/436—Semantic checking
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Stored Programmes (AREA)
Abstract
The invention provides a system and a method for searching and analyzing software defects based on semantic concepts, belonging to the technical field of software engineering. The system and the method comprise a classification module, a semantic concept module and a user feedback module, and realize software defect retrieval and analysis by corresponding steps. The efficiency of retrieval and analysis is improved, so that developers and users can find relevant software defects and appropriate solutions.
Description
Technical Field
The invention relates to a system and a method for searching and analyzing software defects based on semantic concepts, belonging to the technical field of software engineering.
Background
In recent years, a large amount of software has been rapidly developed. In particular, to meet social needs, larger-scale, more sophisticated software has been designed. At the same time, many software bugs also occur. Sometimes, these defects result in costly downtime. It is clear that the damage caused by software defects is immeasurable, so that appropriate software defects must be retrieved and analyzed during the software development and testing phase. To accomplish the task of retrieving and analyzing related software defects, a number of models and methods for software defect exploration and analysis have been proposed. However, they are not accurate for defect retrieval, and developers using defect retrieval need to do more work instead. In addition, these models and methods do not take into account comments from developers and users.
Disclosure of Invention
In order to solve the technical problem that software defect retrieval and analysis at the software development and test stage is lacked in the prior art, the invention provides a system and a method for retrieving and analyzing software defects based on semantic concepts, and the adopted technical scheme is as follows:
a semantic concept based software bug retrieval and analysis system, the system comprising:
a classification module for classifying the tags; wherein the label is a keyword for marking software defects;
a semantic concept module for assisting users and developers to retrieve tag and solution information;
the user feedback module is used for managing and storing user feedback information and recording user information;
the classification module comprises:
the software defect and solution generation module is used for a developer to generate a software defect and a solution corresponding to the software defect;
the marking module is used for marking the software defects and the solutions corresponding to the software defects through a public classification method;
the defect classification module is used for classifying the software defects into different types according to the labels and obtaining defect types;
the semantic concept module comprises:
the tag tree module is used for generating a tag tree added with a semantic concept model by a cosine similarity algorithm; wherein the semantic concept model comprises tags and related semantic concepts, and the semantic concept model is used for describing the relationship between the tags and the related semantic concepts;
a user query module for receiving a user query request;
the label recommending module is used for recommending the label to the user according to the related semantic concept through a query request input by the user;
the user feedback module comprises:
the evaluation module is used for evaluating and grading the result of the user and providing comments;
a feedback information management module for managing user feedback information;
the feedback information storage module is used for storing user feedback information;
and the user information recording module is used for recording the user information of the user.
Further, the system further comprises:
a customer input module for customer input of questions containing label and end face information;
a display module for displaying a software bug report;
the data matching module is used for matching the key words and the end face information which are used for inputting with the existing data in the database; wherein the end face information comprises a category, a language and a defect description;
an association establishing module for associating the tag with the software bug and a solution.
Further, the calculation model of the cosine similarity algorithm is as follows:
wherein, sim (t)i,tj) For similarity values between labels, label ti,tjIs represented as a vector; w is akiIs a label tiAt defect bkWeight of (1), wkiIs defined as TF-IDF and has
A semantic concept based software bug retrieval and analysis method, the method comprising:
a classification step for classifying the labels; wherein the label is a keyword for marking software defects;
a semantic concept step for helping users and developers to retrieve tag and solution information;
a user feedback step for managing and storing user feedback information and recording user information at the same time;
the step of classifying includes:
a software defect and solution generation step for a developer to generate a software defect and a solution corresponding to the software defect;
a marking step for marking the software defect and the solution corresponding to the software defect by a mass classification method;
a defect classification step for classifying the software defects into different types according to the labels and obtaining defect types;
the semantic concept step comprises:
a tag tree step for generating a tag tree to which a semantic concept model is added by a cosine similarity algorithm; wherein the semantic concept model comprises tags and related semantic concepts, and the semantic concept model is used for describing the relationship between the tags and the related semantic concepts;
a user query step for receiving a user query request;
a label recommending step for recommending the label to the user according to the related semantic concept through a query request input by the user;
the user feedback step comprises:
an evaluation step for the user to evaluate, grade and provide comments on the results;
a feedback information management step for managing user feedback information;
a feedback information storage step for storing user feedback information;
a user information recording step for recording user information of the user.
Further, the method further comprises:
a customer input step for customer input of a question containing label and end face information;
a display step for displaying a software defect report;
a data matching step for matching the keyword and the end face information used for inputting with the existing data in the database; wherein the end face information comprises a category, a language and a defect description;
an association establishing step for associating the tag with the software defect and solution.
Further, the calculation model of the cosine similarity algorithm is as follows:
wherein, sim (t)i,tj) For similarity values between labels, label ti,tjIs represented as a vector; w is akiIs a label tiAt defect bkWeight of (1), wkiIs defined as TF-IDF and has
The invention has the beneficial effects that:
the invention provides a software defect retrieval and analysis system and method based on semantic concepts, which are a method for establishing a semantic concept model capable of supporting searching videos by using keywords and semantic concepts based on a complementary classification technology based on user feedback for software retrieval. Developers and users can search for appropriate software defects by entering keywords and performing some correlation operation called an end face. The system and method employs a semantic concept model to improve the accuracy of the search. In addition, this system allows developers and users to submit feedback information in order to meet their needs. The technology combines keyword search, end face search and popular search to retrieve and analyze software defects. Meanwhile, the popular classification method allows a user to mark files according to specific meanings of the files on the internet. Even though mass-taxonomy has a myriad of advantages, its main drawback is the lack of semantic information. Therefore, in order to avoid the defect of the popular classification method, the invention utilizes a semantic concept model to improve the accuracy of semantic analysis, and the method improves the efficiency of retrieval and analysis, so that developers and users can find out relevant software defects and appropriate solutions.
Drawings
FIG. 1 is a schematic diagram of the system of the present invention;
FIG. 2 is a software defect retrieval interface;
FIG. 3 is a label recommendation interface;
FIG. 4 is a result list of software defects;
FIG. 5 is a defect report display interface;
FIG. 6 is a schematic diagram of an example of labeling of the system of the present invention;
FIG. 7 is an architectural diagram of a semantic concept model;
FIG. 8 is a flow chart of the method of the present invention.
Detailed Description
The present invention will be further described with reference to the following specific examples, but the present invention is not limited to these examples.
Example 1:
a semantic concept based software bug retrieval and analysis system, as shown in fig. 1, comprising:
a classification module for classifying the tags; wherein the label is a keyword for marking software defects;
a semantic concept module for assisting users and developers to retrieve tag and solution information;
the user feedback module is used for managing and storing user feedback information and recording user information;
the classification module comprises:
the software defect and solution generation module is used for a developer to generate a software defect and a solution corresponding to the software defect;
the marking module is used for marking the software defects and the solutions corresponding to the software defects through a public classification method;
the defect classification module is used for classifying the software defects into different types according to the labels and obtaining defect types;
the semantic concept module comprises:
the tag tree module is used for generating a tag tree added with a semantic concept model by a cosine similarity algorithm; wherein the semantic concept model comprises tags and related semantic concepts, and the semantic concept model is used for describing the relationship between the tags and the related semantic concepts;
a user query module for receiving a user query request;
the label recommending module is used for recommending the label to the user according to the related semantic concept through a query request input by the user;
the user feedback module comprises:
the evaluation module is used for evaluating and grading the result of the user and providing comments;
a feedback information management module for managing user feedback information;
the feedback information storage module is used for storing user feedback information;
and the user information recording module is used for recording the user information of the user.
Wherein the system further comprises:
a customer input module for customer input of questions containing label and end face information;
a display module for displaying a software bug report;
the data matching module is used for matching the key words and the end face information which are used for inputting with the existing data in the database; wherein the end face information comprises a category, a language and a defect description;
an association establishing module for associating the tag with the software bug and a solution.
The software defect retrieval and analysis system based on semantic concepts provided by the embodiment mainly comprises three modules:
a classification module: this module is used to classify tags. After the developer creates a new software bug report containing a bug description and associated solutions, it is tagged with a label that will be arranged into a bug classification according to a classmark algorithm.
A semantic concept module: this module is the core of the system. In this module, we apply a semantic concept model. The semantic concept model is a hierarchical system, this structure consisting of a concept tree (labels and related semantic concepts). The semantic concept model describes the relationship between the labels and the related semantic concepts to help users and developers retrieve related software defect description and solution information. If the user selects an appropriate semantic concept, our system will present a list of results for the software bug.
A user feedback module: the user can weigh the relevant solutions and give comments after the software defects and the relevant solutions given by the system, and the module manages and stores feedback information of the user and records information of the user so as to improve system performance.
At the beginning of the search process, developers and users can create software bugs and their solutions and mark them by adding tags. Similar tags will be collected by semantic analysis. If the user enters a question containing keywords and end face information, our system will recommend relevant tags. Developers and users may weigh these tags to improve the quality of tag recommendations. Finally, developers can find software bugs and corresponding solutions by proposing recommended tags.
The present embodiment proposes a software defect retrieval and analysis system based on semantic concept, the system includes four aspects of retrieval analysis: keyword and facet retrieval, public analysis, semantic concept model and user feedback. The overall process of analysis of the system's search is: first, taking the input keyword "email management" and related information as an example, fig. 2 shows that a developer inputs the keyword "email management" and related information. When a developer clicks on the query button to search for relevant defects, the system will recommend the appropriate labels and relevant semantic concepts. FIG. 3 shows recommended tags and associated semantic concepts. The developer can select the appropriate tags and concepts. In this example, the developer has selected the label "mail management" and the associated semantic concept "data error" to describe the software bug type of "mail management". FIG. 4 shows a result list of software defects when the developer selects relevant labels and semantic concepts, with 5 defect reports sorted by user's score. Each report contains information about the software bug report, such as program name, type, language, etc. The developer may select any defect report to examine the details of the defect report. In FIG. 5, the software bug report is described in detail to the developer. This report shows the details of the software "Yahoo Test". Wherein, the "user score" represents the average score of the user feedback, and the "solution" represents the debugging method of the defect. In the software bug report, a developer can give a feedback score and related comments from 1 to 5 to improve the quality of software bug retrieval and analysis.
The retrieval and analysis of each aspect is specifically as follows:
and aiming at keyword and facet retrieval:
if the developer only enters the appropriate keywords and facial information, the associated software bug reports will be displayed. If the information input by the developer is the same as the data in the database, the system of the invention directly displays the related software defect report. Table 1 shows a keyword and face information input by a user. This request is described in a piece of XML. In table 1, the keyword elements show keywords input by the user. The facet information includes "kind", "language", and "defect description".
TABLE 1
Table 2 describes the feedback obtained from the queries entered in fig. 3. This feedback includes "software name", "general description", "source code line number", "kind", "defect description", "defect cause", "solution". The core part of this feedback information is "defect description", "defect cause", "solution".
TABLE 2
Aiming at the popular classification method:
the user enters a keyword for marking a defect called a "label". The developer may symbolize a representation of the software bug with some labels. The label classifies the defects into clusters. Fig. 6 shows an example of labeling. In this example, the frequency of marking the defect "slow" with "mail management" is the highest (8 times), and the frequency of marking the defect "test error" with "mail management" is the highest (6 times), so the classification of "slow" and "test error" is the same and has the same label "mail management". Finally, each defect is classified.
For the semantic concept model:
and (3) eliminating tag ambiguity by using a semantic concept model so as to enhance the accuracy of tag recommendation. In the system, the concept of the concept network is used for various common understandings of semantic concepts, so as to describe software defects. So in our system, the semantic concept model consists of clusters of labels and related semantic concepts. Formally, we have the following definitions.
Definition 1: defining a label tree: (i) a label tree is a structure of a semantic concept model. (ii) t1, t2.. tk is a sequence of tags that contain semantic concepts and are on a path from the root to the leaves. (iii) The tag tree consists of tag clusters and related semantic concepts.
Definition 2: to implement a tag cluster, it is necessary to calculate the degree of similarity between tags. In this embodiment, the similarity between the labels is calculated by using a cosine similarity algorithm. The calculation model of the cosine similarity algorithm is as follows:
wherein, sim (t)i,tj) For similarity values between labels, label ti,tjIs represented as a vector; w is akiIs a label tiAt defect bkWeight of (1), wkiIs defined as TF-IDF and has
FIG. 7 is an example of a tree structure describing a semantic concept model showing tag clusters and semantic concepts. For example, the labels "dynamic emulation module" and "embedded SL" are both classified into "simulator". "date errors" and "sequence errors" represent the semantic concept of a "dynamic simulation module". In fact, these semantic concepts explain different drawbacks of the software "dynamic simulation module".
And calculating the similarity of the labels by a cosine similarity algorithm, classifying the labels into clusters and forming a semantic concept model. Table 3 shows a detailed algorithm for semantic understanding. In this algorithm, if the similarity between the label and the original node ti exceeds a threshold, the label becomes a child node of the tree; otherwise, the labels and nodes in the tree are classified into one class. Then, if there is a child node, adding the semantic concept to the child node; otherwise it is added to the parent node. Finally, a semantic concept model is built step by step.
TABLE 3
Feedback for the user:
as described in the previous section, a feedback module is used to allow the user to make appropriate assessments of software errors and provide comments. The feedback reflects the interests of the user. It helps the system to improve the retrieval quality. The feedback information of the user comprises the scores given to the software defects by the user, the average scores of the software defects and the comments of the user.
Example 2
A method for searching and analyzing software defects based on semantic concepts, as shown in fig. 8, the method comprising:
a classification step for classifying the labels; wherein the label is a keyword for marking software defects;
a semantic concept step for helping users and developers to retrieve tag and solution information;
a user feedback step for managing and storing user feedback information and recording user information at the same time;
the step of classifying includes:
a software defect and solution generation step for a developer to generate a software defect and a solution corresponding to the software defect;
a marking step for marking the software defect and the solution corresponding to the software defect by a mass classification method;
a defect classification step for classifying the software defects into different types according to the labels and obtaining defect types;
the semantic concept step comprises:
a tag tree step for generating a tag tree to which a semantic concept model is added by a cosine similarity algorithm; wherein the semantic concept model comprises tags and related semantic concepts, and the semantic concept model is used for describing the relationship between the tags and the related semantic concepts;
a user query step for receiving a user query request;
a label recommending step for recommending the label to the user according to the related semantic concept through a query request input by the user;
the user feedback step comprises:
an evaluation step for the user to evaluate, grade and provide comments on the results;
a feedback information management step for managing user feedback information;
a feedback information storage step for storing user feedback information;
a user information recording step for recording user information of the user.
Wherein the method further comprises:
a customer input step for customer input of a question containing label and end face information;
a display step for displaying a software defect report;
a data matching step for matching the keyword and the end face information used for inputting with the existing data in the database; wherein the end face information comprises a category, a language and a defect description;
an association establishing step for associating the tag with the software defect and solution.
The calculation model of the cosine similarity algorithm is as follows:
wherein, sim (t)i,tj) For similarity values between labels, label ti,tjIs represented as a vector; w is akiIs a label tiAt defect bkWeight of (1), wkiIs defined as TF-IDF and has
At the beginning of a retrieval process, a developer creates a software defect report and a solution, a public classification method is used for marking the defect report created by the developer, the defects are classified into different classes (class 1-class N) according to labels to obtain software defect types, then a cosine similarity algorithm is used for generating a label tree added with a semantic concept model, after a user inputs a query request, the system recommends related labels to the user according to the semantic concept, and the user scores the query after viewing the recommended labels. Feedback is then given to our system.
Although the present invention has been described with reference to the preferred embodiments, it should be understood that various changes and modifications can be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (6)
1. A semantic concept based software bug retrieval and analysis system, the system comprising:
a classification module for classifying the tags; wherein the label is a keyword for marking software defects;
a semantic concept module for assisting users and developers to retrieve tag and solution information;
the user feedback module is used for managing and storing user feedback information and recording user information;
the classification module comprises:
the software defect and solution generation module is used for a developer to generate a software defect and a solution corresponding to the software defect;
the marking module is used for marking the software defects and the solutions corresponding to the software defects through a public classification method;
the defect classification module is used for classifying the software defects into different types according to the labels and obtaining defect types;
the semantic concept module comprises:
the tag tree module is used for generating a tag tree added with a semantic concept model by a cosine similarity algorithm; wherein the semantic concept model comprises tags and related semantic concepts, and the semantic concept model is used for describing the relationship between the tags and the related semantic concepts;
a user query module for receiving a user query request;
the label recommending module is used for recommending the label to the user according to the related semantic concept through a query request input by the user;
the user feedback module comprises:
the evaluation module is used for evaluating and grading the result of the user and providing comments;
a feedback information management module for managing user feedback information;
the feedback information storage module is used for storing user feedback information;
and the user information recording module is used for recording the user information of the user.
2. The software bug retrieval and analysis system of claim 1, further comprising:
a customer input module for customer input of questions containing label and end face information;
a display module for displaying a software bug report;
the data matching module is used for matching the key words and the end face information which are used for inputting with the existing data in the database; wherein the end face information comprises a category, a language and a defect description;
an association establishing module for associating the tag with the software bug and a solution.
3. The software bug retrieval and analysis system of claim 1, wherein the computational model of the cosine similarity algorithm is:
wherein, sim (t)i,tj) For similarity values between labels, label ti,tjIs represented as a vector; w is akiIs a label tiAt defect bkWeight of (1), wkiIs defined as TF-IDF and has
4. A software defect retrieval and analysis method based on semantic concepts, which is characterized by comprising the following steps:
a classification step for classifying the labels; wherein the label is a keyword for marking software defects;
a semantic concept step for helping users and developers to retrieve tag and solution information;
a user feedback step for managing and storing user feedback information and recording user information at the same time;
the step of classifying includes:
a software defect and solution generation step for a developer to generate a software defect and a solution corresponding to the software defect;
a marking step for marking the software defect and the solution corresponding to the software defect by a mass classification method;
a defect classification step for classifying the software defects into different types according to the labels and obtaining defect types;
the semantic concept step comprises:
a label tree step for generating a label tree added with a semantic concept model by a cosine similarity algorithm; wherein the semantic concept model comprises tags and related semantic concepts, and the semantic concept model is used for describing the relationship between the tags and the related semantic concepts;
a user query step for receiving a user query request;
a label recommending step for recommending the label to the user according to the related semantic concept through a query request input by the user;
the user feedback step comprises:
an evaluation step for the user to evaluate, grade and provide comments on the results;
a feedback information management step for managing user feedback information;
a feedback information storage step for storing user feedback information;
a user information recording step for recording user information of the user.
5. The software bug retrieval and analysis method of claim 4, further comprising:
a customer input step for customer input of a question containing label and end face information;
a display step for displaying a software defect report;
a data matching step for matching the keyword and the end face information used for inputting with the existing data in the database; wherein the end face information comprises a category, a language and a defect description;
an association establishing step for associating the tag with the software defect and solution.
6. The software defect retrieval and analysis method of claim 4, wherein the computational model of the cosine similarity algorithm is:
wherein, sim (t)i,tj) For similarity values between labels, label ti,tjIs represented as a vector; w is akiIs a label tiAt defect bkWeight of (1), wkiIs defined as TF-IDF and has
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811285850.5A CN109299381B (en) | 2018-10-31 | 2018-10-31 | Software defect retrieval and analysis system and method based on semantic concept |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811285850.5A CN109299381B (en) | 2018-10-31 | 2018-10-31 | Software defect retrieval and analysis system and method based on semantic concept |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109299381A CN109299381A (en) | 2019-02-01 |
| CN109299381B true CN109299381B (en) | 2020-04-24 |
Family
ID=65145146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811285850.5A Expired - Fee Related CN109299381B (en) | 2018-10-31 | 2018-10-31 | Software defect retrieval and analysis system and method based on semantic concept |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109299381B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113886467B (en) * | 2021-10-25 | 2024-05-14 | 上海航天计算机技术研究所 | Software defect library maintenance method based on keyword extraction |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101814098A (en) * | 2010-05-11 | 2010-08-25 | 天津大学 | Method for obtaining software security defects based on vertical search and semantic annotation |
| CN104951458A (en) * | 2014-03-26 | 2015-09-30 | 华为技术有限公司 | Method and equipment for helping processing based on semantic recognition |
| CN106649557A (en) * | 2016-11-09 | 2017-05-10 | 北京大学（天津滨海）新代信息技术研究院 | Semantic association mining method for defect report and mail list |
| CN107608732A (en) * | 2017-09-13 | 2018-01-19 | 扬州大学 | A kind of bug search localization methods based on bug knowledge mappings |
| CN106844194B (en) * | 2016-12-21 | 2018-06-15 | 北京航空航天大学 | A kind of construction method of multi-level software fault diagnosis expert system |
-
2018
- 2018-10-31 CN CN201811285850.5A patent/CN109299381B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101814098A (en) * | 2010-05-11 | 2010-08-25 | 天津大学 | Method for obtaining software security defects based on vertical search and semantic annotation |
| CN104951458A (en) * | 2014-03-26 | 2015-09-30 | 华为技术有限公司 | Method and equipment for helping processing based on semantic recognition |
| CN106649557A (en) * | 2016-11-09 | 2017-05-10 | 北京大学（天津滨海）新代信息技术研究院 | Semantic association mining method for defect report and mail list |
| CN106844194B (en) * | 2016-12-21 | 2018-06-15 | 北京航空航天大学 | A kind of construction method of multi-level software fault diagnosis expert system |
| CN107608732A (en) * | 2017-09-13 | 2018-01-19 | 扬州大学 | A kind of bug search localization methods based on bug knowledge mappings |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109299381A (en) | 2019-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11500818B2 (en) | Method and system for large scale data curation | |
| Da Costa et al. | A framework for evaluating the results of the szz approach for identifying bug-introducing changes | |
| CN104834729B (en) | Topic recommends method and topic recommendation apparatus | |
| Al-Kofahi et al. | Fuzzy set approach for automatic tagging in evolving software | |
| CN108701254A (en) | System and method for the tracking of dynamic family, reconstruction and life cycle management | |
| US20150293917A1 (en) | Confidence Ranking of Answers Based on Temporal Semantics | |
| CN110968695A (en) | Intelligent labeling method, device and platform based on active learning of weak supervision technology | |
| CN110321291A (en) | Test cases intelligent extraction system and method | |
| Halkidi et al. | Data mining in software engineering | |
| US9760828B2 (en) | Utilizing temporal indicators to weight semantic values | |
| CN111427775B (en) | Method level defect positioning method based on Bert model | |
| Kejriwal et al. | A two-step blocking scheme learner for scalable link discovery. | |
| CN116703328B (en) | Project review method and system | |
| CN116485597B (en) | Standardized training method based on post capability model | |
| US20190325258A1 (en) | Early feedback of schematic correctness in feature management frameworks | |
| Cheng et al. | A similarity integration method based information retrieval and word embedding in bug localization | |
| Arnarsson et al. | Supporting knowledge re-use with effective searches of related engineering documents-a comparison of search engine and natural language processing-based algorithms | |
| CN105511869A (en) | Demand tracking system and method based on user feedback | |
| CN109299381B (en) | Software defect retrieval and analysis system and method based on semantic concept | |
| tong et al. | Mining and analyzing user feedback from app reviews: An econometric approach | |
| Joppi et al. | POP: mining POtential Performance of new fashion products via webly cross-modal query expansion | |
| CN115292167A (en) | Life cycle prediction model construction method, device, equipment and readable storage medium | |
| Wang | Intelligent data mining techniques for automatic service management | |
| Bansal et al. | Literature review of finding duplicate bugs in open source systems | |
| Kangwanwisit et al. | A component recommendation model for issues in software projects |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200424 Termination date: 20211031 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |