CN117076281A - Software quality assessment method based on deep learning - Google Patents
Software quality assessment method based on deep learning Download PDFInfo
- Publication number
- CN117076281A CN117076281A CN202311323934.4A CN202311323934A CN117076281A CN 117076281 A CN117076281 A CN 117076281A CN 202311323934 A CN202311323934 A CN 202311323934A CN 117076281 A CN117076281 A CN 117076281A
- Authority
- CN
- China
- Prior art keywords
- software
- value
- information
- fault
- quality
- 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 20
- 238000011156 evaluation Methods 0.000 claims abstract description 45
- 238000012545 processing Methods 0.000 claims abstract description 44
- 230000000694 effects Effects 0.000 claims abstract description 33
- 238000000034 method Methods 0.000 claims abstract description 33
- 230000008569 process Effects 0.000 claims abstract description 15
- 230000000903 blocking effect Effects 0.000 claims description 7
- 238000012423 maintenance Methods 0.000 claims description 7
- NFEIBWMZVIVJLQ-UHFFFAOYSA-N mexiletine hydrochloride Chemical compound [Cl-].CC([NH3+])COC1=C(C)C=CC=C1C NFEIBWMZVIVJLQ-UHFFFAOYSA-N 0.000 claims description 6
- 230000006870 function Effects 0.000 description 12
- 238000005259 measurement Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013522 software testing Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3438—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment monitoring of user actions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3452—Performance evaluation by statistical analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The invention discloses a software quality assessment method based on deep learning, which relates to the technical field of software quality assessment, and the technical scheme is characterized by comprising the following steps: acquiring basic information when a user uses the software and acquiring fault information of the software in the use process; wherein the basic information comprises katon information, active information and proficiency information; processing and analyzing active information and proficiency information when a user uses software to obtain an easy-to-use value; the easy-to-use value comprises an activity value and a proficiency value, wherein the activity value is a value which is obtained by processing and analyzing the activity information of a user when the user uses the software and used for representing the activity of the software, and the effect is that the quality of the software is evaluated according to the aspects of the click information, the activity information and the proficiency information of the user when the user uses the software and the fault information of the software in the use process, so that the quality of the software is measured from multiple aspects, and a more objective evaluation result can be obtained.
Description
Technical Field
The invention relates to the technical field of software quality evaluation, in particular to a software quality evaluation method based on deep learning.
Background
Currently, quality assessment for software is mainly performed by means of the results of software testing. In the research of the national and international software quality evaluation theory, a plurality of models for evaluating the software quality based on the test result are provided. According to the requirements of the test method, a large number of targeted software tests are used for obtaining test values corresponding to each measurement index, then the measurement values of the measurement elements are calculated according to the measurement method, and finally the software quality is evaluated according to the weights of the measurement elements; however, the existing software quality assessment method based on deep learning cannot analyze and judge the quality of the software according to the basic information of the software in the use process and the fault information of the software in the use process, so that the subsequent upgrading treatment on the quality improvement of the software is inconvenient.
Disclosure of Invention
Aiming at the defects existing in the prior art, the invention aims to provide a software quality assessment method based on deep learning.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a software quality assessment method based on deep learning, the method comprising the steps of:
acquiring basic information when a user uses the software and acquiring fault information of the software in the use process; wherein the basic information includes a cartoon, an active information and a proficiency information.
Processing and analyzing active information and proficiency information when a user uses software to obtain an easy-to-use value; the easy-to-use value comprises an activity value and a proficiency value, wherein the activity value is a value which is obtained by processing and analyzing activity information when a user uses software and used for representing the activity of the software, and the proficiency value is a value which is obtained by processing and analyzing proficiency information when the user uses the software and used for representing the proficiency degree of the user to the software.
The method comprises the steps that the blocking information when a user uses software and fault information of the software in the use process are processed and analyzed to obtain a damage value; the damage value comprises a stuck value and a fault value, wherein the stuck value is a numerical value used for representing software stuck and obtained by processing and analyzing stuck information when a user uses software, and the fault value is a numerical value used for representing software faults and obtained by processing and analyzing fault information of the software in the using process.
And carrying out joint analysis processing on the easy-to-use value and the damage value to obtain an evaluation value.
And evaluating the software quality through the evaluation value.
Preferably, the activity information includes software active person number information and software total daily active time length information.
And processing and marking the information of the number of active people of the software to obtain an active people value N.
And processing and marking the daily active total duration information of the software to obtain a daily active total duration value ZSC of the software.
By an active functionCalculating to obtain an activity value HYZ; wherein a1 and a2 are influence factors and are greater than zero.
Preferably, the proficiency information includes time length information for finding a processing item when the user uses the software.
And (3) carrying out value taking and marking on the time length information of the search processing items when the user uses the software to obtain a proficiency value SLZ.
Preferably by an easy-to-use functionCalculating to obtain an easy-to-use value YYZ; wherein b1 and b2 are influencing factors and are greater than zero.
The jamming information comprises jamming times information and jamming time length information.
And taking the value of the blocking number information and marking to obtain a blocking number value KDC.
And taking and marking the katen time length information to obtain a katen time length KDS.
Through a katon functionCalculating to obtain a katon value KDZ; wherein c1 and c2 are influencing factors and are greater than zero.
Preferably, the fault information includes fault times information and fault duration information.
And taking the value and marking the fault times information to obtain a fault times value GZC.
And taking the value and marking the fault duration information to obtain a fault duration value GZS.
By fault functionCalculating to obtain a fault value GZZ; wherein d1 and d2 are influencing factors and are greater than zero.
Preferably by a processing functionCalculating to obtain SHZ; wherein e1 and e2 are influencing factors and are greater than zero.
By evaluating functionsCalculating to obtain an evaluation value PGZ; wherein f1 and f2 are influence factors and are greater than zero.
Preferably, the evaluation value PGZ is compared with a preset evaluation threshold Q:
and if the evaluation value PGZ is less than or equal to a preset evaluation threshold value Q, the software quality is not qualified.
And if the evaluation value PGZ is larger than a preset evaluation threshold value Q, the software quality is qualified.
Preferably, the length information required for the software to recover to normal and the engineer participation information required for the software to recover to normal are acquired when the user uses the software to get stuck or fails.
Preferably, the software recovery time length value RJH is obtained by processing and marking the time length information required for the software to recover.
And taking the value and marking the information of the number of persons, which is needed by the software to restore, of engineers to obtain the number of persons RSZ needed by the software to restore.
By analysing a functionCalculating to obtain a service value FWZ; where k1 and k2 are influencing factors and are greater than zero.
Preferably, the service value FWZ is compared with a preset service threshold P:
if the service value FWZ is less than or equal to the preset service threshold value P, the software after-sale maintenance service is good.
If the service value FWZ > the preset service threshold value P, the software after-sale maintenance service is poor.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the quality of the software is evaluated from the aspects of the click information, the active information and the proficiency information when the user uses the software and the fault information of the software in the use process, so that the quality of the software is measured from multiple aspects, a more objective evaluation result can be obtained, and the quality of the software is unqualified because the evaluation value PGZ is smaller than a preset evaluation threshold value Q, namely that the software has more click and faults, the user using the software is inactive, and the user becomes difficult to learn when operating the software; under the condition that the evaluation value PGZ is 55, the quality of the software is qualified because the evaluation value PGZ is larger than the preset evaluation threshold value Q, namely that the software is blocked and has fewer faults, the user using the software is active, the user becomes simple and easy to learn when operating the software, and the subsequent software quality is convenient to upgrade under the condition that the quality of the software is poor.
Drawings
Fig. 1 is a schematic flow chart of a software quality evaluation method based on deep learning.
Detailed Description
Reference is made to figure 1.
The embodiment of the invention provides a software quality evaluation method based on deep learning.
A software quality assessment method based on deep learning, the method comprising the steps of:
acquiring basic information when a user uses the software and acquiring fault information of the software in the use process; wherein the basic information includes katon information, active information, and proficiency information.
Processing and analyzing active information and proficiency information when a user uses software to obtain an easy-to-use value; the easy-to-use value comprises an activity value and a proficiency value, wherein the activity value is a value which is obtained by processing and analyzing activity information when a user uses software and used for representing the activity of the software, and the proficiency value is a value which is obtained by processing and analyzing proficiency information when the user uses the software and used for representing the proficiency degree of the user to the software.
The method comprises the steps that the blocking information when a user uses software and fault information of the software in the use process are processed and analyzed to obtain a damage value; the damage value comprises a stuck value and a fault value, the stuck value is a numerical value used for representing software stuck and obtained by processing and analyzing stuck information when a user uses software, and the fault value is a numerical value used for representing software faults and obtained by processing and analyzing fault information of the software in the using process.
And carrying out joint analysis processing on the easy-to-use value and the damage value to obtain an evaluation value.
And evaluating the software quality through the evaluation value.
The invention evaluates the quality of the software from the aspects of the click information, the active information and the proficiency information when the user uses the software and the fault information of the software in the use process, thereby measuring the quality of the software from multiple aspects and obtaining more objective evaluation results.
The activity information includes software activity number information and software total daily activity duration information.
And processing and marking the information of the number of active people of the software to obtain an active people value N.
And processing and marking the daily active total duration information of the software to obtain a daily active total duration value ZSC of the software.
It should be noted that the activity degree of the user when using the software can be reflected by the information of the number of active people of the software and the total daily active time length of each person of the software, and the quality of the software setting can be reflected by the activity degree of the user when using the software.
The value of the active person number N is 1000 if the active person number using the software is 1000, and 10000 if the active person number using the software is 10000.
If the number of active people per day of a certain software is 2, the active time length of one person using the software is 10 minutes, the value ZSC of the active total time length per day of one person is 10, and the active time length of the other person using the software is 30 minutes, and the value ZSC of the active total time length per day of one person is 30.
By an active functionCalculating to obtain an activity value HYZ; wherein a1 and a2 are influence factors and are greater than zero.
It should be noted that here, the value of a1 is set to 0.001, the value of a2 is set to 1,the calculated active average time length of each person of the website is +.>In case the value of (2) is 10, by the active function +.>Calculating to obtain an activity value HYZ of 11, and when the value of the activity number N is 10000,/for the case of the activity number N>In the case of a value of 10, by means of an active functionThe activity value HYZ is calculated to be 110.
The proficiency information includes time information for the user to find a transaction while using the software.
And (3) carrying out value taking and marking on the time length information of the search processing items when the user uses the software to obtain a proficiency value SLZ.
It should be noted that when a certain piece of software is used, after a worker opens the software, the worker needs to search corresponding processing items on the software, if the worker needs to spend more time searching corresponding processing items on the software, the user is unskilled in the software, and the inconvenience of software setting, namely the poor software design quality, is also described; if the staff needs to spend a short time to search the corresponding processing matters on the software, the user is skilled in the software, and the simplicity and convenience of the software setting are also indicated, namely, the software design quality is good.
If the user turns on the software and then finds the relevant processing item after 10 seconds, the value of the proficiency value SLZ is 10, and if the user turns on the software and then finds the relevant processing item after 3 seconds, the value of the proficiency value SLZ is 3.
By easy-to-use functionsCalculating to obtain an easy-to-use value YYZ; wherein b1 and b2 are influencing factors and are greater than zero.
Here, the value of b1 is set to 1, the value of b2 is set to-1, and when the active value HYZ is 110 and the value of the proficiency value SLZ is 3, the function is easily usedCalculating to obtain an easy-to-use value YYZ of 107; when the activity value HYZ is 60 and the proficiency value SLZ is 10, the function is easily usedThe easy-to-use value YYZ was calculated to be 50.
The jamming information comprises jamming times information and jamming time length information.
And taking the value of the blocking number information and marking to obtain a blocking number value KDC.
And taking and marking the katen time length information to obtain a katen time length KDS.
It should be noted that, if the number of times and the duration of the occurrence of the jamming of the software are more, the quality of the software is worse, if the number of times and the duration of the occurrence of the jamming of the software are less, the quality of the software is better, if the number of times of the occurrence of the jamming of the software is 3, the value of the value KDC of the jamming number is 3, and if the number of times of the occurrence of the jamming of the software is 6, the value of the value KDC of the jamming number is 6; if the duration of the software is 10 seconds, the value of the stuck duration value KDS is 10, and if the duration of the software is 6 seconds, the value of the stuck duration value KDS is 6.
Through a katon functionCalculating to obtain a katon value KDZ; wherein c1 and c2 are influencing factors and are greater than zero.
It should be noted that, here, the values of c1 and c2 are 1, and the value of the katon time value KDC is 3, and the value of the katon duration value KDS is 10, by the katon functionCalculating to obtain a katon value KDZ of 13; under the condition that the value of the katon number KDC is 6 and the value of the katon time long value KDS is 10, the katon function is adopted for +.>The calculated stuck value KDZ is 16.
The fault information includes fault number information and fault duration information.
And taking the value and marking the fault times information to obtain a fault times value GZC.
And taking the value and marking the fault duration information to obtain a fault duration value GZS.
It should be noted that, if the number of times of the software failure and the failure time length are more, the quality of the software is worse, if the number of times of the software failure and the failure time length are less, the quality of the software is better, if the number of times of the software failure is 3, the value of the failure time value GZC is 3, and if the number of times of the software failure is 5, the value of the failure time value GZC is 5; if the duration of the software failure is 10 minutes, the value of the failure duration value GZS is 10, and if the duration of the software failure is 20 minutes, the value of the failure duration value GZS is 20.
By fault functionCalculating to obtain a fault value GZZ; wherein d1 and d2 are influencing factors and are greater than zero.
It should be noted that the number of the components,setting the values of d1 and d2 to 1, and passing through a fault function when the value of the fault sub-value GZC is 3 and the value of the fault duration value GZS is 10Calculating to obtain a fault value GZZ of 13; in the case of a fault sub-value GZC of 6 and a fault duration value GZS of 20, the fault function is used +.>The calculated fault value GZZ is 26.
By processing functionsCalculating to obtain SHZ; wherein e1 and e2 are influencing factors and are greater than zero.
It should be noted that here, e1 and e2 are set to 1, and in the case where the katon value KDZ is 16 and the fault value GZZ is 13, the processing function is used to processCalculating to obtain SHZ as 29; in the case of a stuck value KDZ of 10 and a fault value GZZ of 26, by the processing function +.>The SHZ was calculated to be 36.
By evaluating functionsCalculating to obtain an evaluation value PGZ; wherein f1 and f2 are influence factors and are greater than zero.
It should be noted that, here, the value of f1 is set to 1, the value of f2 is set to 100, and when the easy-to-use value YYZ is 50 and the SHZ value is 20, the function is evaluatedCalculating to obtain an evaluation value PGZ of 55; under the condition that the easy-to-use value YYZ is 30 and the SHZ value is 10, the method uses an evaluation function +.>The evaluation value PGZ was calculated to be 40.
Comparing the evaluation value PGZ with a preset evaluation threshold value Q:
and if the evaluation value PGZ is less than or equal to a preset evaluation threshold value Q, the software quality is not qualified.
And if the evaluation value PGZ is larger than a preset evaluation threshold value Q, the software quality is qualified.
It should be noted that, here, the preset evaluation threshold Q is set to 50, and in the case that the evaluation value PGZ is 40, since the evaluation value PGZ is less than the preset evaluation threshold Q, it indicates that the software quality is unqualified, that is, that the software has a lot of jams and faults, and that the user using the software is inactive, and the user becomes difficult to learn when operating the software; under the condition that the evaluation value PGZ is 55, since the evaluation value PGZ is larger than the preset evaluation threshold value Q, the quality of the software is qualified, namely that the software has less jamming and faults, and the user using the software is active, and the user becomes simple and easy to learn when operating the software.
In the second embodiment, the following technical features are added on the basis of the first embodiment:
a software quality assessment method based on deep learning obtains the information of the length required by software restoration and the information of the number of persons involved in engineers required by software restoration when a user uses software to get stuck or failed.
And processing and marking the length information required by the software to recover to be normal to obtain a software recovery time length value RJH.
And taking the value and marking the information of the number of persons, which is needed by the software to restore, of engineers to obtain the number of persons RSZ needed by the software to restore.
It should be noted that when the software is stuck and fails, a software engineer is required to maintain, the subsequent service degree of the software can be reflected by acquiring the length information required by the software to recover to normal and the number of persons involved in the engineer required by the software to recover to normal, if the length of time required by the software to recover to normal is 10 minutes, the length of time value for recovering the software RJH is 10, and if the length of time required by the software to recover to normal is 6 minutes, the length of time value for recovering the software RJH is 6; if the number of the engineers involved in the software recovery is 10, the value of the RSZ required by the software recovery is 10, and if the number of the engineers involved in the software recovery is 8, the value of the RSZ required by the software recovery is 8.
By analysing a functionCalculating to obtain a service value FWZ; where k1 and k2 are influencing factors and are greater than zero.
It should be noted that, here, the value of k1 is set to 1, the value of k2 is set to 10, and when the software recovery period value RJH is 10 and the number of people required for software recovery RSZ is 10, the function is analyzedCalculating to obtain a service value FWZ of 11; under the condition that the software recovery time length value RJH is 6 and the value of the human number value RSZ required by software recovery is 5, the method uses an analysis function +.>The calculated service value FWZ is 8.
Comparing the service value FWZ with a preset service threshold P:
if the service value FWZ is less than or equal to the preset service threshold value P, the software after-sale maintenance service is good.
If the service value FWZ > the preset service threshold value P, the software after-sale maintenance service is poor.
It should be noted that, when the preset service threshold P is set to 9 and the service value FWZ is 8, since the service value FWZ is less than the preset service threshold P, it is indicated that the after-sale maintenance service of the software is good; in the case of a service value FWZ of 11, since the service value FWZ > the preset service threshold P, it is indicated that the software after-sales repair service is poor.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to the present invention may occur to one skilled in the art without departing from the principles of the present invention and are intended to be within the scope of the present invention.
Claims (10)
1. A software quality assessment method based on deep learning is characterized by comprising the following steps:
acquiring basic information when a user uses the software and acquiring fault information of the software in the use process; wherein the basic information comprises katon information, active information and proficiency information;
processing and analyzing active information and proficiency information when a user uses software to obtain an easy-to-use value; the easy-to-use value comprises an activity value and a proficiency value, wherein the activity value is a value which is obtained by processing and analyzing activity information when a user uses software and used for representing the activity of the software, and the proficiency value is a value which is obtained by processing and analyzing proficiency information when the user uses the software and used for representing the proficiency degree of the user on the software;
the method comprises the steps that the blocking information when a user uses software and fault information of the software in the use process are processed and analyzed to obtain a damage value; the damage value comprises a stuck value and a fault value, wherein the stuck value is a numerical value used for representing software stuck and obtained by processing and analyzing stuck information when a user uses software, and the fault value is a numerical value used for representing software faults and obtained by processing and analyzing fault information of the software in the use process;
performing joint analysis processing on the easy-to-use value and the damage value to obtain an evaluation value;
and evaluating the software quality through the evaluation value.
2. The method for evaluating the quality of software based on deep learning according to claim 1, wherein the activity information comprises information of the number of active people in the software and information of the total daily active time length of each person in the software;
processing and marking the information of the number of active people of the software to obtain an active people value N;
processing and marking the daily active total duration information of each person of the software to obtain a daily active total duration value ZSC of each person of the software;
by an active functionCalculating to obtain an activity value HYZ; wherein a1 and a2 are influence factors and are greater than zero.
3. The method for evaluating the quality of software based on deep learning according to claim 2, wherein the proficiency information includes time length information of finding a processing item when a user uses the software;
and (3) carrying out value taking and marking on the time length information of the search processing items when the user uses the software to obtain a proficiency value SLZ.
4. A software quality assessment method according to claim 3, wherein the software quality assessment method is characterized by using an easy-to-use functionCalculating to obtain an easy-to-use value YYZ; wherein b1 and b2 are influencing factors and are greater than zero;
the jamming information comprises jamming times information and jamming time length information;
the stuck time information is valued and marked to obtain a stuck time value KDC;
the method comprises the steps of taking value and marking the katen time length information to obtain a katen time length KDS;
through a katon functionCalculating to obtain a katon value KDZ; wherein c1 and c2 are influencing factors and are greater than zero.
5. The software quality assessment method based on deep learning according to claim 4, wherein the fault information includes fault times information and fault duration information;
the fault times information is valued and marked to obtain a fault times value GZC;
the fault duration information is valued and marked to obtain a fault duration value GZS;
by fault functionCalculating to obtain a fault value GZZ; wherein d1 and d2 are influencing factors and are greater than zero.
6. The method for evaluating software quality based on deep learning of claim 5, wherein the software quality is evaluated by processing functionsCalculating to obtain SHZ; wherein e1 and e2 are influencing factors and are greater than zero;
by evaluating functionsCalculating to obtain an evaluation value PGZ; wherein f1 and f2 are influence factors and are greater than zero.
7. The software quality evaluation method based on deep learning according to claim 6, wherein the evaluation value PGZ is compared with a preset evaluation threshold Q:
if the evaluation value PGZ is less than or equal to a preset evaluation threshold value Q, the software quality is not qualified;
and if the evaluation value PGZ is larger than a preset evaluation threshold value Q, the software quality is qualified.
8. The method for evaluating the quality of software based on deep learning according to claim 7, wherein the length information required for restoring the software and the number of persons involved in the engineer required for restoring the software are acquired when a user uses the software to get stuck or failed.
9. The software quality evaluation method based on deep learning according to claim 8, wherein the software recovery time length value RJH is obtained by processing and marking the software recovery time length information;
taking the value and marking the information of the number of persons, which is needed by the software to restore, of engineers to obtain the number of persons RSZ needed by the software to restore;
by analysing a functionCalculating to obtain a service value FWZ; where k1 and k2 are influencing factors and are greater than zero.
10. The software quality assessment method based on deep learning as claimed in claim 9, wherein the service value FWZ is compared with a preset service threshold P:
if the service value FWZ is less than or equal to a preset service threshold value P, the software after-sale maintenance service is good;
if the service value FWZ > the preset service threshold value P, the software after-sale maintenance service is poor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311323934.4A CN117076281A (en) | 2023-10-13 | 2023-10-13 | Software quality assessment method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311323934.4A CN117076281A (en) | 2023-10-13 | 2023-10-13 | Software quality assessment method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117076281A true CN117076281A (en) | 2023-11-17 |
Family
ID=88706374
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311323934.4A Pending CN117076281A (en) | 2023-10-13 | 2023-10-13 | Software quality assessment method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117076281A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647126A (en) * | 2018-04-27 | 2018-10-12 | 上海交通大学 | A kind of monitoring of mobile applications interim card degree and quantitative estimation method |
| CN110659185A (en) * | 2019-10-09 | 2020-01-07 | 云南电网有限责任公司信息中心 | Mobile APP user experience monitoring method based on real user perception |
| CN114676029A (en) * | 2022-04-02 | 2022-06-28 | 中国民航信息网络股份有限公司 | Software access capability evaluation method, device, system and medium |
| WO2023024316A1 (en) * | 2021-08-24 | 2023-03-02 | 天翼数字生活科技有限公司 | Video quality evaluation method and system based on network health index |
| CN116383048A (en) * | 2023-03-28 | 2023-07-04 | 中信银行股份有限公司 | Software quality information processing method and device |
-
2023
- 2023-10-13 CN CN202311323934.4A patent/CN117076281A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647126A (en) * | 2018-04-27 | 2018-10-12 | 上海交通大学 | A kind of monitoring of mobile applications interim card degree and quantitative estimation method |
| CN110659185A (en) * | 2019-10-09 | 2020-01-07 | 云南电网有限责任公司信息中心 | Mobile APP user experience monitoring method based on real user perception |
| WO2023024316A1 (en) * | 2021-08-24 | 2023-03-02 | 天翼数字生活科技有限公司 | Video quality evaluation method and system based on network health index |
| CN114676029A (en) * | 2022-04-02 | 2022-06-28 | 中国民航信息网络股份有限公司 | Software access capability evaluation method, device, system and medium |
| CN116383048A (en) * | 2023-03-28 | 2023-07-04 | 中信银行股份有限公司 | Software quality information processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106780121B (en) | Power consumption abnormity identification method based on power consumption load mode analysis | |
| WO2021179574A1 (en) | Root cause localization method, device, computer apparatus, and storage medium | |
| CA2226112A1 (en) | Method and system for an architecture based analysis of software quality | |
| CN101706749B (en) | Comprehensive processing method based on software safety defect detection | |
| CN107437113A (en) | A kind of distribution main equipment live detection criterion KBS and its implementation | |
| WO2014173270A1 (en) | Human-machine interface detection method and system | |
| CN113592343A (en) | Fault diagnosis method, device, equipment and storage medium of secondary system | |
| CN102393834B (en) | A kind of method of testing of digital man-machine interface of nuclear power station and system | |
| CN111223564A (en) | Noise hearing loss prediction system based on convolutional neural network | |
| CN112464995A (en) | Power grid distribution transformer fault diagnosis method and system based on decision tree algorithm | |
| CN111126759A (en) | Electric energy meter state evaluation method based on abnormal event fault correlation degree | |
| CN108090515A (en) | A kind of environmental rating appraisal procedure based on data fusion | |
| CN111782532B (en) | Software fault positioning method and system based on network abnormal node analysis | |
| CN111400505A (en) | Method and system for matching fault elimination scheme of power consumption information acquisition system | |
| CN111506635A (en) | System and method for analyzing residential electricity consumption behavior based on autoregressive naive Bayes algorithm | |
| CN111045902A (en) | Pressure testing method and device for server | |
| CN114691521A (en) | Software testing platform based on artificial intelligence | |
| O'Reilly et al. | Revisiting the ROC curve for diagnostic applications with an unbalanced class distribution | |
| CN111553073B (en) | High-reliability electronic product reliability verification method | |
| CN117076281A (en) | Software quality assessment method based on deep learning | |
| CN106909508A (en) | A kind of information system testing results method and system | |
| CN112540912A (en) | Software quality evaluation method and system | |
| Junviani et al. | The role of financial ability as a mediator of financial education and financial satisfaction case study at productive in Makassar city | |
| CN111488284A (en) | Simulation operation active detection method for OpenStack cloud platform | |
| CN109459629A (en) | A kind of recovery capability appraisal procedure based on recovery rate |
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 |