CN114328221A - Cross-project software defect prediction method and system based on feature and instance migration - Google Patents
Cross-project software defect prediction method and system based on feature and instance migration Download PDFInfo
- Publication number
- CN114328221A CN114328221A CN202111627584.1A CN202111627584A CN114328221A CN 114328221 A CN114328221 A CN 114328221A CN 202111627584 A CN202111627584 A CN 202111627584A CN 114328221 A CN114328221 A CN 114328221A
- Authority
- CN
- China
- Prior art keywords
- migration
- software
- project
- feature
- prediction
- 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.)
- Pending
Links
- 238000013508 migration Methods 0.000 title claims abstract description 95
- 230000005012 migration Effects 0.000 title claims abstract description 95
- 230000007547 defect Effects 0.000 title claims abstract description 92
- 238000000034 method Methods 0.000 title claims abstract description 89
- 230000008569 process Effects 0.000 claims abstract description 24
- 230000000694 effects Effects 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000012795 verification Methods 0.000 claims abstract description 14
- 238000005259 measurement Methods 0.000 claims description 26
- 238000012546 transfer Methods 0.000 claims description 26
- 238000004422 calculation algorithm Methods 0.000 claims description 24
- 238000004590 computer program Methods 0.000 claims description 14
- 238000010801 machine learning Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 10
- 230000002068 genetic effect Effects 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 9
- 210000000349 chromosome Anatomy 0.000 claims description 7
- 238000013526 transfer learning Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 5
- 230000007246 mechanism Effects 0.000 claims description 5
- 230000002950 deficient Effects 0.000 claims description 2
- 238000010187 selection method Methods 0.000 abstract description 9
- 238000001914 filtration Methods 0.000 abstract description 8
- 238000010586 diagram Methods 0.000 description 9
- 238000011160 research Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 4
- 238000013145 classification model Methods 0.000 description 3
- 230000035772 mutation Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000002921 genetic algorithm search Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Landscapes
- Stored Programmes (AREA)
Abstract
The invention belongs to the field of software defect prediction, and provides a cross-project software defect prediction method and a system based on feature and instance migration, wherein the method comprises the following steps: in the feature migration stage, a wrapping type feature selection method is adopted, the migration feature set is selected according to the prediction effect of the candidate migration feature subset on the verification set, and the effect is more stable and effective than that of a filtering feature selection method; and in the example migration stage, a cost-sensitive learning mode is adopted to improve the proportion of the defect samples with small quantity and high cost in the training process so as to solve the class imbalance problem of the data set, and finally, a prediction model is constructed by using the data after feature migration and example migration so as to improve the prediction performance of single migration angle modeling.
Description
Technical Field
The invention belongs to the field of software defect prediction, and particularly relates to a cross-project software defect prediction method and system based on feature and instance migration.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Software defect prediction can be divided into same-project software defect prediction and cross-project software defect prediction. The same-project software defect prediction refers to the construction of a defect prediction model by using historical version information of the same software, and the cross-project software defect prediction refers to the construction of a model by using historical data information of other software to perform defect prediction on target software. In practical application, a lot of newly developed software does not have enough historical data information to provide, and the software defect prediction and project software defect prediction are difficult to play roles, so that cross-project software defect prediction becomes a research hotspot in the field. Transfer learning is the main means for solving the problem of cross-project software defect prediction, and comprises feature transfer and instance transfer. The following problems still exist in the current research:
1. in the cross-project software defect prediction research based on feature migration, a filtering feature selection method is mostly used, namely, a statistical method is used for calculating the distribution similarity of features on different projects so as to select a proper migration feature, but the method has a single angle for measuring the quality of the migration feature, and the effect of feature migration cannot be guaranteed.
2. In the cross-project software defect prediction research based on example migration, few researches solve the class imbalance problem of the defect data set, the class imbalance problem affects the training process of the model, the model prediction result is biased to defect-free classes with a large number, and the accuracy of the prediction of the defect module is reduced.
3. In the cross-project software defect prediction research, few researches carry out transfer learning from two aspects of characteristics and examples, and the defect prediction effect cannot be further improved.
Disclosure of Invention
In order to solve at least one technical problem in the background art, the invention provides a cross-project software defect prediction method based on feature migration and instance migration, a wrapping type feature selection method is adopted in a feature migration stage, a migration feature set is selected according to the prediction effect of a candidate migration feature subset on a verification set, and the effect is more stable and effective than that of a filtering feature selection method; and in the example migration stage, a cost-sensitive learning mode is adopted to improve the proportion of the defect samples with small quantity and high cost in the training process so as to solve the class imbalance problem of the data set, and finally, a prediction model is constructed by using the data after feature migration and example migration so as to improve the prediction performance of single migration angle modeling.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a cross-project software defect prediction method based on feature and example migration, which comprises the following steps:
acquiring measurement information of an operand and an operation object in software;
constructing software module measurement elements of a source project and a target project according to measurement information in software;
predicting a software module containing defects according to software module measurement elements of a source project and a target project and a software defect prediction model constructed by machine learning; the process of the software defect prediction model constructed by adopting machine learning comprises the following steps: the method comprises the steps of performing transfer learning from two aspects of characteristics and examples, performing characteristic transfer on a target item from a source item by adopting a characteristic transfer method to obtain characteristics for predicting software defects, performing example transfer on the target item from the source item by adopting an example transfer method, increasing the weight of a defect sample with high prediction error cost in a training process, and finding a software module with defects with the highest probability in the target item based on the source item by combining the characteristic transfer and the example transfer.
A second aspect of the present invention provides a cross project software bug prediction system based on feature and instance migration, comprising:
a software module metric construction module configured to:
acquiring measurement information of an operand and an operation object in software;
constructing software module measurement elements of a source project and a target project according to measurement information in software;
a software defect prediction module configured to: predicting a software module containing defects according to software module measurement elements of a source project and a target project and a software defect prediction model constructed by machine learning; the process of the software defect prediction model constructed by adopting machine learning comprises the following steps: the method comprises the steps of performing transfer learning from two aspects of characteristics and examples, performing characteristic transfer on a target item from a source item by adopting a characteristic transfer method to obtain characteristics for predicting software defects, performing example transfer on the target item from the source item by adopting an example transfer method, increasing the weight of a defect sample with high prediction error cost in a training process, and finding a software module with defects with the highest probability in the target item based on the source item by combining the characteristic transfer and the example transfer.
A third aspect of the invention provides a computer-readable storage medium.
A computer readable storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of the method for cross-project software defect prediction based on feature and instance migration as described above.
A fourth aspect of the invention provides a computer apparatus.
A computer apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the feature and instance migration based cross-project software defect prediction method as described above when executing the program.
Compared with the prior art, the invention has the beneficial effects that:
in the invention, a wrapping type feature method based on a genetic algorithm is used in the feature migration stage, the model prediction effect of the candidate feature subset on the verification set is directly used as an index to search an optimal migration feature set, and compared with a common filtering type feature selection method, the migration effect is more stable and superior; in the example migration stage, a cost sensitive learning mechanism is introduced on the basis of the TrAdaboost algorithm, the weight of a small number of defect samples in the training process is improved, the influence brought by the class imbalance problem is effectively relieved, and the example migration effect is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is an overall flow diagram of a cross-project software bug prediction method based on feature migration and instance migration.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Software defect prediction is a very important and practical research topic in the field of software engineering. The software defect prediction technology can construct a software defect prediction model through the historical data information of the software to predict a new software module, so that an effective test resource allocation scheme is provided for testers, the software test efficiency is improved, and the safety and reliability of the software are further ensured.
Interpretation of terms
Software measurement element: an index or parameter describing a software product or software development process. The software metric is a basis for extracting data from software modules (a software module can be a class, a file or a function or even a package), and plays a key role in data quality.
Example one
The invention aims to solve the problems that in cross-project software defect prediction research, a filtering type feature selection method cannot stably and effectively select proper migration features, the class imbalance problem in instance migration and the prediction effect of a model cannot be further improved from a single angle of features or instances.
As shown in fig. 1, the present embodiment provides a cross-project software defect prediction method based on feature and instance migration, which includes the following steps:
s1, acquiring the measurement information of the operand and the operation object in the software;
the measurement information of the operands and the operation objects in the software comprises measurement information such as the number of all different operands, the number of all different operation objects, the number of all operands in the software module, the number of all operation objects in the module and the like;
s2, constructing software module measurement elements of the source project and the target project according to the measurement information in the software;
wherein, the source item refers to other items, and the target item refers to a software item needing to find defects.
S3, predicting the software module with defects according to the software module measurement elements of the source project and the target project and the software defect prediction model constructed by machine learning; the process of the software defect prediction model constructed by adopting machine learning comprises the following steps:
s31, performing feature migration on the target item from the source item by adopting a genetic algorithm-based wrapping feature migration method to obtain features for predicting software defects, and directly searching an optimal migration feature set by taking model prediction F-measure values of the candidate feature subset on the verification set as indexes;
wherein, the F-measure value is a test index for balancing recall ratio and precision ratio.
It should be noted that, in this embodiment, a wrapping-type feature migration method based on a genetic algorithm is adopted, for example, a filtering-type feature migration method and a feature mapping migration method may also be adopted, and the genetic algorithm is a specific implementation algorithm for implementing the wrapping-type feature migration method, and belongs to an algorithm for searching an optimal solution, which is selected because of simplicity and high efficiency.
The technical effect realized by the technology is as follows: compared with a common filtering type feature selection method, the migration effect is more stable and superior.
S32, carrying out example migration on the target project from the source project by using an example migration method based on cost-sensitive learning, introducing a cost-sensitive learning mechanism, and increasing the weight of the defect samples which occupy less parts in the whole data set by the samples of the whole category in the training process;
it should be noted that, in the embodiment, the example migration method based on cost-sensitive learning may also be an example migration method based on similarity selection and an example migration method based on similarity filtering, for example.
The technical effect of the scheme is that the weight of the defect samples with less number in the training process is improved, the influence brought by the class imbalance problem is effectively relieved, and the effect of instance migration is improved.
And S33, combining the feature migration and the instance migration, and finding the software module with the defects with the highest probability in the target item based on the source item.
The method has the advantages that the two data migration methods are combined, the performance of the cross-project software defect prediction method is further improved, the software defect prediction model can be trained by using the data of other software projects under the condition that the historical information of the target project is lack, the software modules with defects at large probability can be effectively found out, reasonable test resource allocation suggestions are provided for testers, and the software test efficiency is greatly improved.
The genetic algorithm-based wrapping type feature migration method comprises the following steps:
s311, defining chromosomes of a genetic algorithm: the genetic algorithm searches for the optimal set of migration features, each chromosome representing a set of candidate features. Each chromosome is represented by a vector containing only 0 and 1. Assuming that the software defect data set has n features, each chromosome can be represented as a vector with the length of n, if the ith dimension data takes a value of 1, the corresponding ith feature is selected into the candidate feature set, and if the ith dimension data takes a value of 0, the corresponding ith feature is not selected.
S312, population initialization: and customizing the initial scale of the population according to the time cost and the coverage range of the characteristic space, wherein chromosomes in the population are generated by random bit strings.
S313, constructing a fitness function: and constructing a verification set through a minimum part of marked target project metric elements, searching an optimal migration feature set by taking an F-measure value obtained by a candidate feature set on the verification set as fitness, and adding a defective sample in a semi-supervised learning algorithm Tri-training prediction result into the verification set in order to further expand and improve the effect of the verification set.
S314, calculating a selection operator: roulette is used to decide which individuals in the population can enter the next generation to generate a new population.
S315, calculating a crossover operator: and carrying out a Crossover operation by adopting a Partial-Mapped crossbar (Partial mapping Crossover operator) operator.
S316, calculating a mutation operator: and carrying out mutation operation by adopting a uniform mutation operator.
The example migration method based on the cost-sensitive learning comprises the following steps:
the example migration method based on the example migration algorithm TrAdaboost (semi-supervised integrated learning algorithm) is characterized in that a plurality of models are trained and are used in an integrated mode, and the t-th classification model refers to the middle model.
A cost sensitive learning mechanism is added into the TrAdaboost algorithm, specifically, an example weight updating strategy of the TrAdaboost algorithm is changed, and different cost sensitive factors are added to each example, so that the weight of a defect sample with high prediction error cost is increased quickly and reduced slowly in the training process.
The weight calculation formula after the change is:
wherein,
represents the weight of the ith sample in the t round, betatWeight coefficient, h, representing classification model of the t-th roundt(x) A prediction label representing the t-th classification model for instance x, c (x) a true label representing instance x,
and expressing a cost adjusting function, wherein the calculation formula of the cost adjusting function is as follows:
wherein, ciRepresenting the cost factor of instance i.
In summary, by means of the above technical solution of the present invention, a genetic algorithm-based wrapping type feature method is used in the feature migration stage, and the model prediction effect of the candidate feature subset on the verification set is directly used as an index to search for the optimal migration feature set, so that the migration effect is more stable and superior than that of a common filtering type feature selection method; a cost sensitive learning mechanism is introduced on the basis of the TrAdaboost algorithm in the instance migration stage, the weight of a small number of defect samples in the training process is improved, the influence brought by the class imbalance problem is effectively relieved, and the instance migration effect is improved.
Example two
The embodiment provides a cross-project software defect prediction system based on feature and instance migration, which comprises:
a software module metric construction module configured to:
acquiring measurement information of an operand and an operation object in software;
constructing software module measurement elements of a source project and a target project according to measurement information in software;
a software defect prediction module configured to: predicting a software module containing defects according to software module measurement elements of a source project and a target project and a software defect prediction model constructed by machine learning; the process of the software defect prediction model constructed by adopting machine learning comprises the following steps: performing migration learning from the aspects of characteristics and examples, performing characteristic migration on a target item from a source item by adopting a characteristic migration method, and selecting a migration characteristic set according to the prediction effect of a candidate migration characteristic subset on a verification set; and (4) carrying out example migration on the target item from the source item by adopting an example migration method, and increasing the weight of the defect sample in the training process.
EXAMPLE III
The present embodiments provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps in the cross-project software defect prediction method based on feature and instance migration as described above.
Example four
The embodiment provides a computer device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of the cross-project software defect prediction method based on feature and example migration.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The cross-project software defect prediction method based on feature and instance migration is characterized by comprising the following steps of:
acquiring measurement information of an operand and an operation object in software;
constructing software module measurement elements of a source project and a target project according to measurement information in software;
predicting a software module containing defects according to software module measurement elements of a source project and a target project and a software defect prediction model constructed by machine learning; the process of the software defect prediction model constructed by adopting machine learning comprises the following steps: the method comprises the steps of performing transfer learning from two aspects of characteristics and examples, performing characteristic transfer on a target item from a source item by adopting a characteristic transfer method to obtain characteristics for predicting software defects, performing example transfer on the target item from the source item by adopting an example transfer method, increasing the weight of a defect sample with high prediction error cost in a training process, and finding a software module with defects with the highest probability in the target item based on the source item by combining the characteristic transfer and the example transfer.
2. The method for predicting defects of cross-project software based on feature and instance migration according to claim 1, wherein the feature migration is performed by a wrapping feature migration method based on a genetic algorithm, and the method comprises the following steps:
defining chromosomes of a genetic algorithm, each chromosome representing a set of candidate features;
customizing the initial scale of the population according to the time cost and the coverage range of the characteristic space;
constructing a fitness function, constructing a verification set through a marked target item metric element, and searching an optimal migration feature set by taking an F-measure value acquired by a candidate feature set on the verification set as the fitness;
roulette is used to determine which individuals in the population can enter the next generation to generate a new population.
3. The method of claim 2, wherein when constructing the fitness function, the method adds the defective samples in the prediction results of the semi-supervised learning algorithm Tri-training into the verification set.
4. The method for predicting the software defects of the cross-project based on the characteristics and the example migration as claimed in claim 1, wherein in the process of performing the example migration on the target project from the source project by adopting the example migration method, the example migration method based on the example migration algorithm TrAdaboost algorithm is adopted to train a multi-round model, and finally the multi-round model is used in an integrated mode.
5. The method for predicting the defect of the cross-project software based on the feature and the example migration as claimed in claim 1, wherein a cost sensitive learning mechanism is added into the TrAdaboost algorithm, an example weight updating strategy of the TrAdaboost algorithm is changed, different cost sensitive factors are added to each example, and the weight of a defect sample with high prediction error cost is increased.
6. The method of claim 1, wherein the metric information of operands and operands in the software comprises metric information of the number of all different operands, the number of all operands in a software module, and the number of all operands in a module.
7. The method for predicting the software defects of the cross-project based on the characteristic and the example migration according to claim 1, wherein in the method for performing the characteristic migration on the target project from the source project by adopting the wrapping type characteristic migration method based on the genetic algorithm, the optimal migration characteristic set is directly searched by taking the model prediction effect of the candidate characteristic subset on the verification set as an index.
8. A cross-project software defect prediction system based on feature and instance migration is characterized by comprising the following steps:
a software module metric construction module configured to:
acquiring measurement information of an operand and an operation object in software;
constructing software module measurement elements of a source project and a target project according to measurement information in software;
a software defect prediction module configured to: predicting a software module containing defects according to software module measurement elements of a source project and a target project and a software defect prediction model constructed by machine learning; the process of the software defect prediction model constructed by adopting machine learning comprises the following steps: the method comprises the steps of performing transfer learning from two aspects of characteristics and examples, performing characteristic transfer on a target item from a source item by adopting a characteristic transfer method to obtain characteristics for predicting software defects, performing example transfer on the target item from the source item by adopting an example transfer method, increasing the weight of a defect sample with high prediction error cost in a training process, and finding a software module with defects with the highest probability in the target item based on the source item by combining the characteristic transfer and the example transfer.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for cross-project software defect prediction based on feature and instance migration according to any one of claims 1 to 7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps in the feature and instance migration based cross-project software defect prediction method of any one of claims 1-7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111627584.1A CN114328221A (en) | 2021-12-28 | 2021-12-28 | Cross-project software defect prediction method and system based on feature and instance migration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111627584.1A CN114328221A (en) | 2021-12-28 | 2021-12-28 | Cross-project software defect prediction method and system based on feature and instance migration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114328221A true CN114328221A (en) | 2022-04-12 |
Family
ID=81015770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111627584.1A Pending CN114328221A (en) | 2021-12-28 | 2021-12-28 | Cross-project software defect prediction method and system based on feature and instance migration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114328221A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115269377A (en) * | 2022-06-23 | 2022-11-01 | 南通大学 | Cross-project software defect prediction method based on optimization instance selection |
-
2021
- 2021-12-28 CN CN202111627584.1A patent/CN114328221A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115269377A (en) * | 2022-06-23 | 2022-11-01 | 南通大学 | Cross-project software defect prediction method based on optimization instance selection |
| CN115269377B (en) * | 2022-06-23 | 2023-07-11 | 南通大学 | Cross-project software defect prediction method based on optimization instance selection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hu et al. | HiSCF: leveraging higher-order structures for clustering analysis in biological networks | |
| CN110659207B (en) | Heterogeneous cross-project software defect prediction method based on nuclear spectrum mapping migration integration | |
| Hesami et al. | Machine learning: its challenges and opportunities in plant system biology | |
| CN107203467A (en) | The reference test method and device of supervised learning algorithm under a kind of distributed environment | |
| He et al. | Evolutionary graph clustering for protein complex identification | |
| CN107844414A (en) | A kind of spanned item mesh based on defect report analysis, parallelization defect positioning method | |
| CN108491228A (en) | A kind of binary vulnerability Code Clones detection method and system | |
| CN113127342B (en) | Defect prediction method and device based on power grid information system feature selection | |
| CN111310860A (en) | Method and computer-readable storage medium for improving performance of gradient boosting decision trees | |
| Yu et al. | Melo: Enhancing model editing with neuron-indexed dynamic lora | |
| Yao et al. | Accurately detecting protein complexes by graph embedding and combining functions with interactions | |
| Ponte-Fernández et al. | Evaluation of existing methods for high-order epistasis detection | |
| CN114706780B (en) | Software defect prediction method based on Stacking integrated learning | |
| CN114781688A (en) | Method, device, equipment and storage medium for identifying abnormal data of business expansion project | |
| CN116993548A (en) | Incremental learning-based education training institution credit assessment method and system for LightGBM-SVM | |
| CN114328221A (en) | Cross-project software defect prediction method and system based on feature and instance migration | |
| Zhou et al. | Deeptle: Learning code-level features to predict code performance before it runs | |
| Lim et al. | Accelerated quality-diversity through massive parallelism | |
| CN117453503A (en) | Optimization method, device, equipment and medium of deep learning operator program | |
| CN116756662A (en) | Yield prediction method and system for optimizing random forest based on Harris eagle algorithm | |
| Nikolikj et al. | Sensitivity Analysis of RF+ clust for Leave-one-problem-out Performance Prediction | |
| CN107291722B (en) | Descriptor classification method and device | |
| CN111026661B (en) | Comprehensive testing method and system for software usability | |
| CN115239122A (en) | Digital power grid software project tester recommendation method and device | |
| Karthika et al. | Genetic Algorithm-Based Feature Selection and Self-Organizing Auto-Encoder (Soae) for Snp Genomics Data Classifications |
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 |