CN117667697A - Software quality assessment method based on deep learning - Google Patents
Software quality assessment method based on deep learning Download PDFInfo
- Publication number
- CN117667697A CN117667697A CN202311661267.0A CN202311661267A CN117667697A CN 117667697 A CN117667697 A CN 117667697A CN 202311661267 A CN202311661267 A CN 202311661267A CN 117667697 A CN117667697 A CN 117667697A
- Authority
- CN
- China
- Prior art keywords
- model
- data
- quality
- software
- test
- 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
- 238000013135 deep learning Methods 0.000 title claims abstract description 23
- 238000012360 testing method Methods 0.000 claims abstract description 86
- 238000000034 method Methods 0.000 claims abstract description 69
- 238000012549 training Methods 0.000 claims abstract description 28
- 238000013441 quality evaluation Methods 0.000 claims abstract description 22
- 238000002372 labelling Methods 0.000 claims abstract description 21
- 238000003062 neural network model Methods 0.000 claims abstract description 18
- 238000011156 evaluation Methods 0.000 claims abstract description 17
- 230000006872 improvement Effects 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 28
- 239000013598 vector Substances 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 14
- 238000005516 engineering process Methods 0.000 claims description 13
- 238000004140 cleaning Methods 0.000 claims description 12
- 125000004122 cyclic group Chemical group 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 11
- 238000012795 verification Methods 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 5
- 238000001303 quality assessment method Methods 0.000 claims description 4
- 238000013138 pruning Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000004321 preservation Methods 0.000 claims description 2
- 238000007418 data mining Methods 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 238000013179 statistical model Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Abstract
The invention discloses a software quality assessment method based on deep learning, which comprises the following steps: the software product is preprocessed, so that subsequent training and testing are facilitated; constructing a deep neural network model, and predicting the quality index of the software product according to the characteristics and the labeling data of the software product; testing the software product, and verifying the prediction capability and accuracy of the model by using the test cases and the test results; and carrying out quality evaluation on the software product according to the prediction result and the test result of the model, and giving corresponding suggestions and improvement measures. The software quality evaluation method based on deep learning has the advantages of higher evaluation efficiency and accuracy, lower manual intervention and cost and suitability for various software.
Description
Technical Field
The invention relates to the technical field of software engineering, in particular to a software quality assessment method based on deep learning.
Background
Software quality assessment is the process of assessing and measuring a software product or system to determine the reliability, availability, efficiency, security of the software and whether it meets the intended functional and performance requirements. Software quality assessment involves the use of a series of assessment methods and tools to determine whether software meets standards and specifications and is compatible with particular user requirements.
The existing software quality evaluation method is mainly divided into a method based on manual review and a method based on data mining, and the method based on data mining is mainly divided into a method based on a statistical model and a method based on a machine learning model, and the two methods have defects and shortcomings: the method based on manual review requires a large number of professionals to evaluate, has high labor cost and low efficiency, is greatly influenced by human factors, and has poor unrepeatability; the method based on the statistical model in the data mining method requires a large amount of parameter adjustment and feature selection, has small data size, low quality, complex data structure and difficult labeling, is sensitive to noise and abnormal values, requires a large amount of training data and calculation resources, has difficult parameter and feature selection, is sensitive to complexity and nonlinearity, and has poor generalization capability and interpretation.
Therefore, it is necessary to provide a new software quality assessment method based on deep learning to solve the above technical problems.
Disclosure of Invention
The invention aims to provide a software quality assessment method based on deep learning, which improves the efficiency and level of software quality management by automatically assessing and analyzing aspects of functions, performances, reliability and the like of software by using a deep learning model.
According to the embodiment of the invention, an intelligent video compression method and system based on video technology comprise the following steps:
s1, data cleaning: preprocessing the original data, and dividing the data set into a training set, a verification set and a test set so as to establish, evaluate and optimize models on different data;
as a further aspect of the present invention, the preprocessing operation specifically includes:
data cleaning: removing invalid or error data, filling missing values or abnormal values, converting data formats or types and the like from the original data so as to improve the data quality;
feature extraction: performing operations such as dimension reduction or feature selection on the original data to extract the features related to or important to the quality index;
labeling: and labeling or classifying the original data to give a quality index value or category corresponding to each sample or each category.
As a further aspect of the present invention, the data segmentation method includes:
and (3) randomly dividing: according to certain rules or criteria, the whole data set is randomly divided into three parts, each part is called a subset;
orderly dividing: the entire data set is divided in order into three parts, each part being called a subset, according to a certain order or condition.
S2, constructing a model: constructing a deep neural network model according to the feature vector obtained by the feature extraction component and the label vector obtained by the labeling component, and selecting proper parameters such as a network structure, a loss function, an optimizer and the like;
as a further scheme of the invention, the deep neural network construction specifically comprises convolution operation, an activation function and a cyclic neural network, wherein the cyclic neural network consists of a plurality of cyclic units, and each cyclic unit comprises a gating unit and an activation unit.
S3, model test: testing a deep neural network model according to the trained model parameters, and evaluating the prediction capability and accuracy of the model by using proper test cases and test results;
as a further aspect of the present invention, the model training includes:
and (3) loading data: reading a data set from a storage medium to a memory, and preprocessing and dividing the data set;
model preservation: and saving the trained model parameters into a file.
As a further scheme of the invention, the problems of data quality, calculation resources and learning rate encountered in model training are as follows:
data quality:
(1) Dividing and grouping data, and dividing a data set according to different categories or attributes so as to facilitate learning and prediction of a model;
(2) The data is enhanced and expanded, and the diversity and complexity of the data are increased by adding random noise, disturbance, transformation and other operations;
computing resources:
(1) Compressing and optimizing the model, pruning, quantifying and distilling by using a technology or a method, and reducing the parameter quantity and the calculation complexity of the model;
(2) Parallel and distributed training of the model, the training process of the model being dispersed over a plurality of devices or nodes by using techniques or methods;
(3) Parallel and distributed testing is carried out on the model, and the testing process of the model is dispersed to a plurality of devices or nodes by using a technology or a method;
learning rate:
(1) The learning rate is adjusted and optimized, and the learning rate is adjusted according to different conditions and targets by using a technology or a method;
(2) Monitoring and evaluating the learning rate, monitoring the variation condition of the learning rate at different points by using a technology or a method, and adjusting the learning rate according to feedback information;
(3) And verifying and testing the learning rate, continuously verifying the performance of the model on the verification set in the training process, adjusting the learning rate according to the verification result, continuously testing the performance of the model on the test set in the testing process, and adjusting the learning rate according to the test result.
S4, quality evaluation: and evaluating the quality of the software product according to the model prediction result and the test result obtained by the model test part, and giving out corresponding suggestions and improvement measures.
As a further aspect of the present invention, the S4 specifically includes:
s41, model prediction results: predicting a true or expected quality level represented by the sample or class in the quality assessment based on the trained model parameters;
s42, test results: evaluating a true or expected quality level represented by the software product in the quality evaluation based on the test cases and test results;
s43, a quality evaluation method: depending on the type and scale of software products, as well as on the quality index and evaluation criteria, a suitable quality assessment method is selected.
As a further scheme of the invention, the quality evaluation method is a fuzzy comprehensive evaluation method, and specifically comprises the following steps:
(1) Establishing a factor set and an evaluation set of comprehensive evaluation;
(2) Factor weight vectors and membership functions are determined.
As a further aspect of the present invention, the system specifically includes:
the data cleaning module is used for removing invalid or erroneous data, filling missing values or abnormal values, converting data formats or types and the like from the original data so as to improve the data quality;
the feature extraction module is used for performing operations such as dimension reduction or feature selection on the original data so as to extract features related to or important to the quality index;
the labeling module is used for labeling or classifying the original data to give a quality index value or category corresponding to the sample or category;
the model construction module is used for constructing a deep neural network model according to the feature vector obtained by the feature extraction component and the label vector obtained by the labeling component, and selecting proper parameters such as a network structure, a loss function, an optimizer and the like;
the model training module is used for training the deep neural network model according to the model parameters obtained by the model construction part and updating the model parameters by using proper parameters such as a loss function, an optimizer and the like;
the model test module is used for testing the deep neural network model according to the trained model parameters and evaluating the prediction capability and accuracy of the model by using proper test cases and test results;
and the quality evaluation module is used for evaluating the quality of the software product according to the model prediction result and the test result obtained by the model test component and giving corresponding suggestions and improvement measures.
The invention provides a software quality assessment method based on deep learning, which has the beneficial effects that: the efficiency and the accuracy of software quality evaluation can be effectively improved, and meanwhile, the manual intervention and the cost are reduced; the method can adapt to software products of different types and scales, and supports various quality indexes and evaluation standards; quality indexes such as defect rate, reliability, safety and the like of the software product can be predicted according to the characteristics and the labeling data of the software product, so that timely and accurate quality feedback and suggestion are provided for a software developer; the software product can be tested by using the deep learning model, and the prediction capability and accuracy of the model are verified by using the test cases and the test results, so that reliable and effective test results are provided for a software developer; the quality evaluation method can evaluate the quality of the software product according to the prediction result and the test result of the model, and give corresponding suggestions and improvement measures, thereby providing valuable and meaningful quality evaluation reports for software developers.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a flow chart of the steps of a software quality evaluation method based on deep learning according to the present invention;
FIG. 2 is a block diagram of a system module of a software quality evaluation method based on deep learning according to the present invention;
fig. 3 is a flowchart of an implementation of a software quality evaluation method based on deep learning according to the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Referring to fig. 1 to fig. 3, a software quality evaluation method based on deep learning provided by an embodiment of the present invention includes:
s1, data cleaning: preprocessing the original data, and dividing the data set into a training set, a verification set and a test set so as to establish, evaluate and optimize models on different data;
in this embodiment, the common methods for data preprocessing include discretization, normalization, standardization, encoding and dimension reduction, and the two methods for data segmentation are random division and ordered division, and the specific formulas related to the corresponding system modules for data cleaning include:
removing invalid or erroneous data:
filling up missing values:
s2, constructing a model: constructing a deep neural network model according to the feature vector obtained by the feature extraction component and the label vector obtained by the labeling component, and selecting proper parameters such as a network structure, a loss function, an optimizer and the like;
in this embodiment, the construction of the deep neural network model specifically includes:
(1) Convolution operation, the calculation formula is:
input-output relationship: the input data is X, the output data is Y, the convolution kernel is W, and the step length is s, and then the method comprises the following steps:
where N is the batch size, C is the channel number, and b is the offset term.
Output dimension: the output data is Y, the convolution kernel is W, the step length is s, and the following steps are:
wherein,is the coefficient corresponding to the step size.
Parameter number: the input data is X, the output data is Y, the convolution kernel is W, and the step length is s, and then the method comprises the following steps:
wherein,is the coefficient corresponding to the step length;
(2) Activating a function, wherein a calculation formula is as follows;
sigmoid function:
tanh function:
ReLU function:
f(x)＝max(0,x)
softmax function:
(3) The cyclic neural network consists of a plurality of cyclic units, and the calculation formula of the cyclic units is as follows:
let the input sequence be x 1 ,x 2 ,…,x n ,x 0 ,x 1 ,…,x n …, there are:
h t ＝f(W h x t +b h )
wherein,
h t ＝h t-1 (g t (W g h t-1 (x t )+b g ))
s3, model test: testing a deep neural network model according to the trained model parameters, and evaluating the prediction capability and accuracy of the model by using proper test cases and test results;
in this embodiment, the problems of data quality, computing resources and learning rate encountered in model training are as follows:
data quality:
(1) Dividing and grouping data, and dividing a data set according to different categories or attributes so as to facilitate learning and prediction of a model;
(2) The data is enhanced and expanded, and the diversity and complexity of the data are increased by adding random noise, disturbance, transformation and other operations;
computing resources:
(1) Compressing and optimizing the model, pruning, quantifying and distilling by using a technology or a method, and reducing the parameter quantity and the calculation complexity of the model;
(2) Parallel and distributed training of the model, the training process of the model being dispersed over a plurality of devices or nodes by using techniques or methods;
(3) Parallel and distributed testing is carried out on the model, and the testing process of the model is dispersed to a plurality of devices or nodes by using a technology or a method;
learning rate:
(1) The learning rate is adjusted and optimized, and the learning rate is adjusted according to different conditions and targets by using a technology or a method;
(2) Monitoring and evaluating the learning rate, monitoring the variation condition of the learning rate at different points by using a technology or a method, and adjusting the learning rate according to feedback information;
(3) And verifying and testing the learning rate, continuously verifying the performance of the model on the verification set in the training process, adjusting the learning rate according to the verification result, continuously testing the performance of the model on the test set in the testing process, and adjusting the learning rate according to the test result.
S4, quality evaluation: performing quality evaluation on the software product according to the model prediction result and the test result obtained by the model test component, and giving out corresponding suggestions and improvement measures;
in this embodiment, the quality evaluation method is a fuzzy comprehensive evaluation method, and specifically includes:
(1) And establishing a factor set and an evaluation set of the comprehensive evaluation. The factor set is a common set composed of elements of various factors affecting the evaluation object, and is generally denoted by U, and n factors: u rm = U 1 ,u 2 ,cdots,u m . An evaluation set is a set of various results that an evaluator may make on an evaluation object, and is generally denoted by V, and m comments: v rm = V 1 ,v 2 ,cdots,v m 。
(2) Factor weight vectors and membership functions are determined. Giving each factor a weight a i The modular fuzzy set of the weight set of each factor is denoted by a: arm=a 1 ,a 2 ,cdots,a m . In the absence of data, we can determine weights by analytic hierarchy process; when data is available, the weights may be determined by an entropy weighting method.
A system for running a deep learning based software quality assessment method, comprising in particular:
the data cleaning module is used for removing invalid or erroneous data, filling missing values or abnormal values, converting data formats or types and the like from the original data so as to improve the data quality;
the feature extraction module is used for performing operations such as dimension reduction or feature selection on the original data so as to extract features related to or important to the quality index;
the labeling module is used for labeling or classifying the original data to give a quality index value or category corresponding to the sample or category;
the model construction module is used for constructing a deep neural network model according to the feature vector obtained by the feature extraction component and the label vector obtained by the labeling component, and selecting proper parameters such as a network structure, a loss function, an optimizer and the like;
the model training module is used for training the deep neural network model according to the model parameters obtained by the model construction part and updating the model parameters by using proper parameters such as a loss function, an optimizer and the like;
the model test module is used for testing the deep neural network model according to the trained model parameters and evaluating the prediction capability and accuracy of the model by using proper test cases and test results;
and the quality evaluation module is used for evaluating the quality of the software product according to the model prediction result and the test result obtained by the model test component and giving corresponding suggestions and improvement measures.
Examples:
taking shopping software as an example:
and a data cleaning module: and cleaning the original data of the shopping software to remove invalid or erroneous data, such as deleting repeated order records, repairing the date with wrong format, and the like, so as to improve the data quality.
And the feature extraction module is used for: and extracting the characteristics of the cleaned data, and selecting proper characteristics to perform dimension reduction or characteristic selection operation according to the requirements and quality indexes of shopping software, such as extracting user behavior characteristics, commodity attribute characteristics and the like.
And the marking module is used for: the data after cleaning and feature extraction is labeled or classified according to the quality index of the shopping software, for example, the order data is labeled with quality labels, such as "high quality", "low quality" or classified into specific categories.
Model construction module: and constructing a deep neural network model according to the feature vector obtained by the feature extraction component and the label vector obtained by the labeling component, and selecting proper network structures such as parameters of a convolutional neural network, a cyclic neural network and the like, a loss function, an optimizer and the like to establish a quality evaluation model.
Model training module: model parameters obtained by the model construction part are used for model training of training data of shopping software, and parameters such as a proper loss function, an optimizer and the like are used for updating the model parameters so as to improve the performance and accuracy of the model.
Model test module: according to the trained model parameters, the model test is carried out by using the test data of the shopping software, and the prediction capability and accuracy of the model are evaluated by using proper test cases and test results, for example, the quality of the shopping software is evaluated by calculating indexes such as accuracy, recall rate and the like.
The quality evaluation module: and carrying out quality evaluation on the shopping software according to the model prediction result and the test result obtained by the model test component, and giving corresponding suggestions and improvement measures, such as giving abnormality detection and error repair suggestions for low-quality orders in the shopping software, optimizing commodity recommendation algorithms and the like.
Compared with the related art, the software quality evaluation method based on deep learning has the following beneficial effects:
the invention provides a software quality assessment method based on deep learning, which can effectively improve the efficiency and accuracy of software quality assessment and simultaneously reduce the manual intervention and cost; the method can adapt to software products of different types and scales, and supports various quality indexes and evaluation standards; quality indexes such as defect rate, reliability, safety and the like of the software product can be predicted according to the characteristics and the labeling data of the software product, so that timely and accurate quality feedback and suggestion are provided for a software developer; the software product can be tested by using the deep learning model, and the prediction capability and accuracy of the model are verified by using the test cases and the test results, so that reliable and effective test results are provided for a software developer; the quality evaluation method can evaluate the quality of the software product according to the prediction result and the test result of the model, and give corresponding suggestions and improvement measures, thereby providing valuable and meaningful quality evaluation reports for software developers.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.
Claims (9)
1. A software quality assessment method based on deep learning, comprising:
s1, data cleaning: preprocessing the original data, and dividing the data set into a training set, a verification set and a test set so as to establish, evaluate and optimize models on different data;
s2, constructing a model: constructing a deep neural network model according to the feature vector obtained by the feature extraction component and the label vector obtained by the labeling component, and selecting proper parameters such as a network structure, a loss function, an optimizer and the like;
s3, model test: testing a deep neural network model according to model parameters trained by the model, and evaluating the prediction capability and accuracy of the model by using proper test cases and test results;
s4, quality evaluation: and evaluating the quality of the software product according to the model prediction result and the test result obtained by the model test part, and giving out corresponding suggestions and improvement measures.
2. The deep learning-based software quality assessment method according to claim 1, wherein the preprocessing operation specifically comprises:
data cleaning: removing invalid or error data, filling missing values or abnormal values, converting data formats or types and the like from the original data to improve the data quality, wherein the formula is expressed as follows;
filling up missing values:
feature extraction: performing operations such as dimension reduction or feature selection on the original data to extract the features related to or important to the quality index;
labeling: and labeling or classifying the original data to give a quality index value or category corresponding to each sample or each category.
3. The deep learning-based software quality assessment method according to claim 1, wherein the deep neural network construction specifically comprises convolution operation, an activation function and a cyclic neural network, the cyclic neural network is composed of a plurality of cyclic units, each cyclic unit comprises a gating unit and an activation unit, and functions commonly used by the activation function are as follows:
sigmoid function:
tanh function:
ReLU function f (x) =max (0, x)
softmax function:
4. the deep learning based software quality assessment method of claim 1, wherein the model training comprises:
and (3) loading data: reading a data set from a storage medium to a memory, and preprocessing and dividing the data set;
model preservation: and saving the trained model parameters into a file.
5. The method for evaluating software quality based on deep learning according to claim 1, wherein,
the step S4 specifically comprises the following steps:
s41, model prediction results: predicting a true or expected quality level represented by the sample or class in the quality assessment based on the trained model parameters;
s42, test results: evaluating a true or expected quality level represented by the software product in the quality evaluation based on the test cases and test results;
s43, a quality evaluation method: depending on the type and scale of software products, as well as on the quality index and evaluation criteria, a suitable quality assessment method is selected.
6. The software quality assessment method based on deep learning according to claim 1, wherein the data segmentation method is as follows:
and (3) randomly dividing: according to certain rules or criteria, the whole data set is randomly divided into three parts, each part is called a subset;
orderly dividing: the entire data set is divided in order into three parts, each part being called a subset, according to a certain order or condition.
7. The deep learning-based software quality assessment method according to claim 4, wherein the problems of data quality, computing resources and learning rate encountered in the model training are as follows:
data quality:
(1) Dividing and grouping data, and dividing a data set according to different categories or attributes so as to facilitate learning and prediction of a model;
(2) The data is enhanced and expanded, and the diversity and complexity of the data are increased by adding random noise, disturbance, transformation and other operations;
computing resources:
(1) Compressing and optimizing the model, pruning, quantifying and distilling by using a technology or a method, and reducing the parameter quantity and the calculation complexity of the model;
(2) Parallel and distributed training of the model, the training process of the model being dispersed over a plurality of devices or nodes by using techniques or methods;
(3) Parallel and distributed testing is carried out on the model, and the testing process of the model is dispersed to a plurality of devices or nodes by using a technology or a method;
learning rate:
(1) The learning rate is adjusted and optimized, and the learning rate is adjusted according to different conditions and targets by using a technology or a method;
(2) Monitoring and evaluating the learning rate, monitoring the variation condition of the learning rate at different points by using a technology or a method, and adjusting the learning rate according to feedback information;
(3) And verifying and testing the learning rate, continuously verifying the performance of the model on the verification set in the training process, adjusting the learning rate according to the verification result, continuously testing the performance of the model on the test set in the testing process, and adjusting the learning rate according to the test result.
8. The software quality evaluation method based on deep learning according to claim 1, wherein the quality evaluation method is a fuzzy comprehensive evaluation method, and specifically comprises:
(1) Establishing a factor set and an evaluation set of comprehensive evaluation;
(2) Factor weight vectors and membership functions are determined.
9. System for operating the method according to any of the preceding claims 1-8, characterized in that the system comprises in particular:
the data cleaning module is used for removing invalid or erroneous data, filling missing values or abnormal values, converting data formats or types and the like from the original data so as to improve the data quality;
the feature extraction module is used for performing operations such as dimension reduction or feature selection on the original data so as to extract features related to or important to the quality index;
the labeling module is used for labeling or classifying the original data to give a quality index value or category corresponding to the sample or category;
the model construction module is used for constructing a deep neural network model according to the feature vector obtained by the feature extraction component and the label vector obtained by the labeling component, and selecting proper parameters such as a network structure, a loss function, an optimizer and the like;
the model training module is used for training the deep neural network model according to the model parameters obtained by the model construction part and updating the model parameters by using proper parameters such as a loss function, an optimizer and the like;
the model test module is used for testing the deep neural network model according to the trained model parameters and evaluating the prediction capability and accuracy of the model by using proper test cases and test results;
and the quality evaluation module is used for evaluating the quality of the software product according to the model prediction result and the test result obtained by the model test component and giving corresponding suggestions and improvement measures.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311661267.0A CN117667697A (en) | 2023-12-06 | 2023-12-06 | Software quality assessment method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311661267.0A CN117667697A (en) | 2023-12-06 | 2023-12-06 | Software quality assessment method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117667697A true CN117667697A (en) | 2024-03-08 |
Family
ID=90066564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311661267.0A Pending CN117667697A (en) | 2023-12-06 | 2023-12-06 | Software quality assessment method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117667697A (en) |
-
2023
- 2023-12-06 CN CN202311661267.0A patent/CN117667697A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111460728B (en) | Method and device for predicting residual life of industrial equipment, storage medium and equipment | |
| CN113407694B (en) | Method, device and related equipment for detecting ambiguity of customer service robot knowledge base | |
| CN116450399B (en) | Fault diagnosis and root cause positioning method for micro service system | |
| CN111343147B (en) | Network attack detection device and method based on deep learning | |
| CN110706039A (en) | Electric vehicle residual value rate evaluation system, method, equipment and medium | |
| CN117034143B (en) | Distributed system fault diagnosis method and device based on machine learning | |
| CN115795351A (en) | Elevator big data risk early warning method based on residual error network and 2D feature representation | |
| CN112508456A (en) | Food safety risk assessment method, system, computer equipment and storage medium | |
| CN112527572A (en) | Disk failure prediction method and device, computer readable storage medium and server | |
| CN117194163A (en) | Computer equipment, fault detection system, method and readable storage medium | |
| Bond et al. | A hybrid learning approach to prognostics and health management applied to military ground vehicles using time-series and maintenance event data | |
| CN117667697A (en) | Software quality assessment method based on deep learning | |
| CN116910526A (en) | Model training method, device, communication equipment and readable storage medium | |
| CN115587333A (en) | Failure analysis fault point prediction method and system based on multi-classification model | |
| CN114757495A (en) | Membership value quantitative evaluation method based on logistic regression | |
| CN115081514A (en) | Industrial equipment fault identification method under data imbalance condition | |
| CN109978038B (en) | Cluster abnormity judgment method and device | |
| CN113591396A (en) | Power grid component fault diagnosis method based on naive Bayesian network | |
| CN113935413A (en) | Distribution network wave recording file waveform identification method based on convolutional neural network | |
| CN117726240B (en) | Quality evaluation classification method and system based on convolutional neural network | |
| CN117539920B (en) | Data query method and system based on real estate transaction multidimensional data | |
| CN117290796A (en) | Bearing fault diagnosis method based on multi-teacher knowledge distillation | |
| CN118034733A (en) | Automobile fault detection model updating method, device, equipment and storage medium | |
| CN117312947A (en) | Cold water machine fault detection method based on capsule network multi-classification model | |
| CN117951028A (en) | Open source software reliability prediction method, device, medium and product |
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 |