CN105653721A - Test effect evaluation method and system - Google Patents
Test effect evaluation method and system Download PDFInfo
- Publication number
- CN105653721A CN105653721A CN201610013131.2A CN201610013131A CN105653721A CN 105653721 A CN105653721 A CN 105653721A CN 201610013131 A CN201610013131 A CN 201610013131A CN 105653721 A CN105653721 A CN 105653721A
- Authority
- CN
- China
- Prior art keywords
- estimator
- evaluation
- sample
- variance
- scoring
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/48—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
Abstract
The invention relates to the technical field of product evaluation and discloses a test effect evaluation method and system. According to the test effect evaluation method and system, evaluation score values of two or more evaluation samples about a target object and a comparison object by two or more evaluators are obtained; difference value analysis is conducted on the evaluation score values of the target object and the comparison object by the evaluators; statistics is conducted on the difference value analysis result, and an evaluation conclusion is obtained. Compared with the prior art, the evaluation contribution of the evaluators or the evaluation samples to the target object or the evaluation object can be judged according to the evaluation score values, and accordingly the evaluation effect can be improved according to the evaluation contribution of the evaluators or the evaluation samples so that subjective evaluation can be more objective.
Description
Technical field
The present invention relates to product evaluation technical field, test effect evaluation method and system particularly to one.
Background technology
Along with the development of science and technology, the electronic product such as smart mobile phone has become instrument indispensable in people's daily life. Further, along with the continuous improvement in audio frequency, video performance performance such as smart mobile phone, it has increasingly becomed important amusement equipment. People can enjoy high-quality music by smart mobile phone, watches high definition picture, video etc. by smart mobile phone. Along with the aggravation of the market competition, increasing manufacturer is using the Consumer's Experience of the product most important target as exploitation new product.
Consumer's Experience for the audio frequency of product, video etc. needs to be evaluated by personnel, and this evaluation often subjectivity is extremely strong, for the confirmation of evaluation object effect, it is difficult to accomplish just right objective evaluation.
Summary of the invention
It is an object of the invention to provide a kind of test effect evaluation method and system so that subjective assessment is more objective.
For solving above-mentioned technical problem, embodiments of the present invention provide a kind of test effect evaluation method, comprise the steps of M estimator of acquisition and about destination object and contrast the object score value for N number of evaluation sample; Wherein, described M, N are the natural number more than 2; Respectively the score value of destination object and contrast object is carried out differential analysis by described M estimator; The result of differential analysis is added up, draws evaluation conclusion.
Embodiments of the present invention additionally provide a kind of test effect evaluation system, comprise: acquisition module, difference calculating module and evaluation module; Described acquisition module about destination object and contrasts the object score value for N number of evaluation sample for obtaining M estimator; Wherein, described M, N are the natural number more than 2; Described difference calculating module is for carrying out differential analysis to described M estimator to the score value of destination object and contrast object respectively; Described evaluation module, for the result of differential analysis is added up, draws evaluation conclusion.
Embodiment of the present invention is in terms of existing technologies, obtain more than 2 estimators and about destination object and contrast object for more than 2 score values evaluating sample, respectively the score value of destination object and contrast object is carried out differential analysis by each estimator, and the result of differential analysis is added up, draw evaluation conclusion. It is thus possible to judge estimator according to score value or evaluate the sample evaluation contribution to destination object and contrast object, and then evaluation effect can be improved according to the evaluation contribution of estimator or evaluation sample, improve the objectivity of subjective assessment.
Preferably, before described M estimator of acquisition is about destination object and the contrast object step for the score value of N number of evaluation sample, also comprises the steps of and be pre-created estimator and evaluate sample database;M estimator is obtained about, in destination object and the contrast object step for the score value of N number of evaluation sample, choosing M estimator and N number of evaluation sample from described estimator and evaluation sample database described; After the described the step respectively score value of destination object and contrast object being carried out differential analysis by described M estimator, also comprise the steps of the described estimator of renewal and evaluate sample database. It is thus possible to improve estimator and evaluate the objective evaluation ability of sample.
Preferably, in the step of the described estimator of described renewal and evaluation sample database, comprise following sub-step: each estimator is carried out variance analysis; Wherein, the scoring variance of described estimator is that each estimator about destination object and contrasts the object scoring variance for described N number of evaluation sample; Scoring variance according to M estimator is changed scoring variance in described M estimator and is met the first pre-conditioned estimator. It is thus possible to improve the objective evaluation ability that estimator is overall.
Preferably, in the step of the described estimator of described renewal and evaluation sample database, comprise following sub-step: each is evaluated sample and carries out variance analysis; Wherein, the scoring variance of described evaluation sample is that each evaluation sample about destination object and contrasts the object scoring variance for described M estimator; Scoring variance according to N number of evaluation sample is changed scoring variance yields in described N number of evaluation sample and is met the second pre-conditioned evaluation sample. The objective evaluation effect that sample is overall is evaluated it is thus possible to improve.
Preferably, the described first pre-conditioned scoring variance for described estimator is minimum less than the scoring variance of a preset value or described estimator; The described second pre-conditioned scoring variance for described evaluation sample is minimum less than the scoring variance of a preset value or described evaluation sample. So that estimator and evaluation sample have preferably objective evaluation ability.
Accompanying drawing explanation
Fig. 1 is the flow chart testing effect evaluation method according to first embodiment of the invention;
Fig. 2 is the structured flowchart testing effect evaluation system according to second embodiment of the invention.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, the embodiments of the present invention are explained in detail. But, it will be understood by those skilled in the art that in each embodiment of the present invention, propose many ins and outs in order to make reader be more fully understood that the application. But, even without these ins and outs with based on the many variations of following embodiment and amendment, it is also possible to realize the application each claim technical scheme required for protection.
First embodiment of the present invention relates to a kind of test effect evaluation method. Idiographic flow is as it is shown in figure 1, test effect evaluation method comprises the steps of
Step 101: be pre-created estimator and evaluate sample database.
Estimator such as can choose randomly from target customers, evaluates sample and can choose randomly from the various resources that can obtain, and certainly, evaluating sample database needs long-term and Dynamic Maintenance renewal. Namely the estimator chosen and evaluation sample constitute estimator and evaluate the data base of sample. Present embodiment is not specifically limited for the creation method of estimator and evaluation sample database.
Step 102: choose M estimator and N number of evaluation sample from estimator and evaluation sample database.Wherein, M, N are the natural number more than 2. That is, from estimator storehouse, select a number of estimator randomly, and select a number of evaluation sample from evaluating sample database. Wherein, the number of estimator is preferably greater than or equal to 5, evaluates sample and is preferably greater than or equal to 10.
Present embodiment is described in detail for smart mobile phone speaker yupin effect subjective assessment. 6 estimators are randomly selected from estimator storehouse. Evaluate sample to choose from evaluation sample database. Wherein, evaluating sample can be that the song 20 that on the widely used the Internet of user, download rate is more is first, it is possible to comprise the tinkle of bells, song, light music etc., makes the form variation as far as possible of sample.
Step 103: obtain M estimator and about destination object and contrast the object score value for N number of evaluation sample.
Destination object is desirable to carry out the model machine of effect judgement, contrasts liking target effect or rival's like product. By contrasting Recognition Different, provide direction thus improving for product. Same song is copied in two model machines, and by volume adjusting to maximum, outside carrying out, puts broadcasting. The tonequality difference of mobile phone is best embodied when volume is maximum.
Play the song 1 of two model machines successively, play and marked by estimator. All song playing duration can be all 20 seconds, by that analogy, records 6 estimators subjective judgment score for all songs, and wherein, scoring divides evaluation with 10-0 from excellent to bad, rounds. 6 estimators are about shown in the score value record such as table () of 20 first songs.
Step 104: respectively the score value of destination object and contrast object is carried out differential analysis by M estimator.
For 6 estimators to target machine (destination object) and contrast machine (contrast object) differential analysis result such as table (two) shown in.
Table (two)
Being better than contrast machine on the occasion of representing target machine effect, negative value represents target machine effect and is inferior to contrast machine.
Step 105: the result of differential analysis is added up, draws evaluation conclusion.
Being added up by the differential analysis result in his-and-hers watches (two), statistics draws on the occasion of 52 times, and negative value 8 times, therefore conclusion can determine whether to be better than contrast machine for target machine.
Step 106: update estimator and evaluate sample database.
Specifically, update estimator and evaluation sample database specifically comprise following sub-step 1060 to sub-step 1063:
Sub-step 1060: each estimator is carried out variance analysis. Wherein, the scoring variance of estimator is that each estimator about destination object and contrasts the object scoring variance for N number of evaluation sample. In present embodiment, the scoring variance of every estimator is about scoring variance for 20 first songs of target machine and contrast machine.
Calculating the scoring variance of 6 estimators successively, the computational methods of the scoring variance of each estimator are as follows:
Assuming that the scoring difference of two model machines of song 1 is x1 by estimator 1, the scoring difference to song 2 is x2, and the scoring difference to song 3 is x3 ..., by that analogy.
First calculate this estimator for all evaluating the meansigma methods of the scoring difference of sample (song), here calculate 20 data, namely n=20 (song 1 of estimator 1, song 2 ...)
Then calculate variance according still further to formula and calculate the variance of all estimators, altogether obtaining 6 data, in the most footline of Table (two):
Sub-step 1061: change scoring variance in M estimator according to the scoring variance of M estimator and meet the first pre-conditioned estimator.It is preferred that the first pre-conditioned scoring variance for estimator is minimum less than the scoring variance of a preset value or estimator. Variance is more little, and it is more invalid to contrast, it was shown that estimator is minimum to differentiation contribution.
Similarly, it is also possible to carry out variance analysis to evaluating sample, the scoring variance more New Appraisement Sample Storehouse according to N number of evaluation sample. Specifically, sub-step 1062 to sub-step 1063 is comprised:
Sub-step 1062: each is evaluated sample and carries out variance analysis. Wherein, the scoring variance evaluating sample is that each evaluation sample about destination object and contrasts the object scoring variance for M estimator. Mode similar to the above is used to calculate variance, it is assumed that estimator 1 is y1 in the scoring difference of two model machines by song 1, and the scoring difference to estimator 2 is y2, and the scoring difference to estimator 3 is y3 ..., by that analogy. Calculate meansigma methods and variance equally. It it is 6 the difference is that every first song variance calculates collecting sample number, always have 20 data, in the most terminal column of Table (two): what horizontal variance embodied is the dispersion degree of same song different evaluation person's difference, can be understood as song diversity on two model machines, variance is more little, and to represent differentiation more little, namely evaluates sample more little for the contribution evaluated.
Sub-step 1063: change scoring variance yields in N number of evaluation sample according to the scoring variance of N number of evaluation sample and meet the second pre-conditioned evaluation sample. It is preferred that the second pre-conditioned scoring variance for evaluating sample is minimum less than the scoring variance of a preset value or evaluation sample. Therefore, it can the obsolescence principle minimum according to variance. Song 10, song 11, song 20 variance is all 0.14, it is proposed that eliminates and changes, and this kind of song is minimum to differentiation contribution.
The application of test effect evaluation method is illustrated by present embodiment again through a concrete example. Evaluation task: the display effect evaluation of target machine and contrast machine. From estimator storehouse, select 5 estimators, Sample Storehouse is selected 10 pictures or video as evaluating sample from evaluating. Target machine and contrast machine carry out display effect contrast, 5 estimators successively 10 display effects evaluating sample is marked. Scoring record is as shown in table (three).
Table (three)
Respectively the score value of target machine and contrast machine is carried out differential analysis by 5 estimators, shown in differential analysis result such as table (four). Pass through differential analysis, it is easy to show that target machine is due to contrast machine.
Table (four)
| Estimator 1 | Estimator 2 | Estimator 3 | Estimator 4 | Estimator 5 | Picture variance | |
| Picture 1 | 1 | 0 | 0 | 0 | 0 | 0.16 |
| Picture 2 | 1 | 0 | 0 | 1 | 1 | 0.24 |
| Picture 3 | 1 | 1 | 0 | 2 | 1 | 0.4 |
| Picture 4 | 2 | 0 | 1 | 2 | 2 | 0.64 |
| Picture 5 | -1 | 1 | 0 | 0 | 1 | 0.56 |
| Picture 6 | 0 | 0 | 0 | 1 | 1 | 0.24 |
| Picture 7 | 0 | 1 | 0 | 0 | 1 | 0.24 |
| Picture 8 | 2 | 2 | 0 | 1 | 1 | 0.56 |
| Picture 9 | 0 | 0 | 0 | 0 | 3 | 1.44 |
| Picture 10 | 0 | 0 | 0 | 1 | 0 | 0.16 |
| Estimator's variance | 0.84 | 0.45 | 0.09 | 0.56 | 0.69 |
Then the scoring variance of 5 estimators is calculated again respectively, most footline in table (four), and 10 scoring variances evaluating samples, terminal column as most in table (four). Therefrom can drawing, judge 3 variance is minimum, it is necessary to eliminating, the variance of picture 1 and picture 10 is minimum, it is necessary to eliminate. Therefore, present embodiment passes through data-driven, test system proposes the mode of Continuous optimization, accomplishes more objective rational judged result.
Present embodiment is in terms of existing technologies, obtain M estimator and about destination object and contrast the object score value for N number of evaluation sample, and respectively the score value of destination object and contrast object is carried out differential analysis by M estimator, result according to differential analysis determines evaluation conclusion, and eliminate the estimator minimum for weak effect alienation contribution and evaluate sample, so that subjective assessment more objectivity.
Additionally, what deserves to be explained is, the test effect evaluation method of present embodiment is not limited only to the evaluation of above-mentioned audio frequency or display effect, can be also used for other subjective evaluations, such as, abnormal smells from the patient, taste, sense of touch etc., the present invention should not be exemplified as limit with above-mentioned. In addition, the method of present embodiment is except carrying out test effect assessment, can be also used for estimator and evaluation sample are evaluated, by Variance Analysis Evaluation person and evaluation sample, it is possible to eliminate the estimator that evaluation result contribution degree is minimum or evaluate sample.
The step of various methods divides above, is intended merely to description clear, it is achieved time can be merged into a step or some step is split, and is decomposed into multiple step, as long as comprising identical logical relation, all in the protection domain of this patent; To adding inessential amendment in algorithm or in flow process or introducing inessential design, but do not change the core design of its algorithm and flow process all in the protection domain of this patent.
Second embodiment of the present invention relates to a kind of test effect evaluation system, as in figure 2 it is shown, comprise: creation module, acquisition module, difference calculating module, evaluation module and more new module.
Creation module is used for being pre-created estimator and evaluating sample database, acquisition module for from estimator with evaluate sample database and choose M estimator and N number of evaluation sample, and obtain M estimator about destination object with contrast the object score value for N number of evaluation sample. M, N are the natural number more than 2, it is preferred that, M more than or equal to 5, N more than or equal to 10.
Difference calculating module is for carrying out differential analysis to M estimator to the score value of destination object and contrast object respectively. Evaluation module, for the result of differential analysis is added up, draws evaluation conclusion.
More new module updates estimator for the result according to differential analysis and evaluates sample database. Specifically, more new module comprises: estimator's analysis module and estimator eliminate module. Estimator's analysis module is for carrying out variance analysis to each estimator. The scoring variance of estimator is that each estimator about destination object and contrasts the object scoring variance for described N number of evaluation sample. Estimator eliminates module and meets the first pre-conditioned estimator for scoring variance in scoring variance M the estimator of replacing according to M estimator.
More new module also comprises: estimator's analysis module and estimator eliminate module. Evaluate sample analysis module and evaluate the superseded module of sample. Evaluate sample analysis module and carry out variance analysis for each is evaluated sample. The scoring variance evaluating sample is that each evaluation sample about destination object and contrasts the object scoring variance for M estimator. The first pre-conditioned scoring variance for estimator is minimum less than the scoring variance of a preset value or estimator. Evaluate the superseded module of sample and meet the second pre-conditioned evaluation sample for scoring variance yields in the scoring variance N number of evaluation sample of replacing according to N number of evaluation sample; The second pre-conditioned scoring variance for evaluating sample is minimum less than the scoring variance of a preset value or evaluation sample.
It should be noted that present embodiment can adopt the score value of scoring device collection estimator, and carry out differential analysis by the processor score value to collecting, and then quickly draw evaluation result. Simultaneously, estimator and evaluation sample can also be carried out variance analysis by processor respectively, calculate each estimator and evaluate the sample contribution degree for evaluation result, and eliminate according to default obsolescence principle according to the size of contribution degree and to be unsatisfactory for pre-conditioned estimator and to evaluate sample, so that subjective evaluation more objectivity.
It is seen that, present embodiment is the system embodiment corresponding with the first embodiment, and present embodiment can be worked in coordination enforcement with the first embodiment.The relevant technical details mentioned in first embodiment is still effective in the present embodiment, in order to reduce repetition, repeats no more here. Correspondingly, the relevant technical details mentioned in present embodiment is also applicable in the first embodiment.
It is noted that each module involved in present embodiment is logic module, in actual applications, a logical block can be a physical location, it is also possible to be a part for a physical location, it is also possible to realize with the combination of multiple physical locations. Additionally, for the innovative part highlighting the present invention, do not introduced by the unit less close with solving technical problem relation proposed by the invention in present embodiment, but this is not intended that in present embodiment to be absent from other unit.
It will be understood by those skilled in the art that the respective embodiments described above are to realize specific embodiments of the invention, and in actual applications, it is possible in the form and details it is done various change, without departing from the spirit and scope of the present invention.
Claims (10)
1. a test effect evaluation method, it is characterised in that comprise the steps of
Obtain M estimator and about destination object and contrast the object score value for N number of evaluation sample; Wherein, described M, N are the natural number more than 2;
Respectively the score value of destination object and contrast object is carried out differential analysis by described M estimator;
The result of differential analysis is added up, draws evaluation conclusion.
2. test effect evaluation method according to claim 1, it is characterised in that before described M estimator of acquisition is about destination object and the contrast object step for the score value of N number of evaluation sample, also comprise the steps of
It is pre-created estimator and evaluates sample database;
M estimator is obtained about, in destination object and the contrast object step for the score value of N number of evaluation sample, choosing M estimator and N number of evaluation sample from described estimator and evaluation sample database described;
After the described the step respectively score value of destination object and contrast object being carried out differential analysis by described M estimator, also comprise the steps of
Update described estimator and evaluate sample database.
3. test effect evaluation method according to claim 2, it is characterised in that in the step of the described estimator of described renewal and evaluation sample database, comprises following sub-step:
Each estimator is carried out variance analysis; Wherein, the scoring variance of described estimator is that each estimator about destination object and contrasts the object scoring variance for described N number of evaluation sample;
Scoring variance according to M estimator is changed scoring variance in described M estimator and is met the first pre-conditioned estimator.
4. the test effect evaluation method according to Claims 2 or 3, it is characterised in that in the step of the described estimator of described renewal and evaluation sample database, comprises following sub-step:
Each is evaluated sample and carries out variance analysis; Wherein, the scoring variance of described evaluation sample is that each evaluation sample about destination object and contrasts the object scoring variance for described M estimator;
Scoring variance according to N number of evaluation sample is changed scoring variance yields in described N number of evaluation sample and is met the second pre-conditioned evaluation sample.
5. test effect evaluation method according to claim 4, it is characterised in that the described first pre-conditioned scoring variance for described estimator is minimum less than the scoring variance of a preset value or described estimator;
The described second pre-conditioned scoring variance for described evaluation sample is minimum less than the scoring variance of a preset value or described evaluation sample.
6. test effect evaluation method according to claim 1, it is characterised in that described M is more than or equal to 5; Described N is more than or equal to 10.
7. a test effect evaluation system, it is characterised in that comprise: acquisition module, difference calculating module and evaluation module;
Described acquisition module about destination object and contrasts the object score value for N number of evaluation sample for obtaining M estimator; Wherein, described M, N are the natural number more than 2;
Described difference calculating module is for carrying out differential analysis to described M estimator to the score value of destination object and contrast object respectively;
Described evaluation module, for the result of differential analysis is added up, draws evaluation conclusion.
8. test effect evaluation system according to claim 7, it is characterised in that also comprise creation module and more new module;
Described creation module is used for being pre-created estimator and evaluating sample database;
Described more new module updates described estimator for the result according to described differential analysis and evaluates sample database.
9. test effect evaluation system according to claim 8, it is characterised in that described more new module comprises: estimator's analysis module and estimator eliminate module;
Described estimator's analysis module is for carrying out variance analysis to each estimator; Wherein, the scoring variance of estimator is that each estimator about destination object and contrasts the object scoring variance for described N number of evaluation sample;
Described estimator eliminates module and meets the first pre-conditioned estimator for scoring variance in scoring variance described M the estimator of replacing according to M estimator.
10. test effect evaluation system according to claim 9, it is characterised in that described more new module also comprises: evaluate sample analysis module and evaluate the superseded module of sample;
Described evaluation sample analysis module carries out variance analysis for each is evaluated sample; Wherein, the scoring variance evaluating sample is that each evaluation sample about destination object and contrasts the object scoring variance for described M estimator; The described first pre-conditioned scoring variance for described estimator is minimum less than the scoring variance of a preset value or described estimator;
Described evaluation sample is eliminated module and is met the second pre-conditioned evaluation sample for scoring variance yields in the scoring variance described N number of evaluation sample of replacing according to N number of evaluation sample; The described second pre-conditioned scoring variance for described evaluation sample is minimum less than the scoring variance of a preset value or described evaluation sample.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610013131.2A CN105653721A (en) | 2016-01-10 | 2016-01-10 | Test effect evaluation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610013131.2A CN105653721A (en) | 2016-01-10 | 2016-01-10 | Test effect evaluation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105653721A true CN105653721A (en) | 2016-06-08 |
Family
ID=56484182
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610013131.2A Pending CN105653721A (en) | 2016-01-10 | 2016-01-10 | Test effect evaluation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105653721A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110314023A1 (en) * | 2010-06-17 | 2011-12-22 | Microsoft Corporation | Online Stratified Sampling for Classifier Evaluation |
| CN103119584A (en) * | 2010-12-17 | 2013-05-22 | 北京交通大学 | Machine translation evaluation device and method |
| CN104573747A (en) * | 2013-10-17 | 2015-04-29 | 北大方正集团有限公司 | Character evaluation method and device |
-
2016
- 2016-01-10 CN CN201610013131.2A patent/CN105653721A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110314023A1 (en) * | 2010-06-17 | 2011-12-22 | Microsoft Corporation | Online Stratified Sampling for Classifier Evaluation |
| CN103119584A (en) * | 2010-12-17 | 2013-05-22 | 北京交通大学 | Machine translation evaluation device and method |
| CN104573747A (en) * | 2013-10-17 | 2015-04-29 | 北大方正集团有限公司 | Character evaluation method and device |
Non-Patent Citations (1)
| Title |
|---|
| 马春燕: "用LONGFORD方法对主观评分中的异常分数进行分数调整", 《中国考试》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2950551A1 (en) | Method for recommending multimedia resource and apparatus thereof | |
| CN104699711B (en) | A kind of recommended method and server | |
| CN105205180B (en) | A kind of evaluation method and device of knowledge mapping | |
| CN104282322B (en) | A kind of mobile terminal and its method and apparatus for identifying song climax parts | |
| CN104021162B (en) | A kind of method and device given a mark for multimedia resource | |
| CN106055617A (en) | Data pushing method and device | |
| CN109948068A (en) | A kind of recommended method and device of interest point information | |
| CN103970806B (en) | Method and device for establishing lyric emotion classification model | |
| CN105915960A (en) | User type determination method and device | |
| CN107992595A (en) | A kind of learning Content recommends method, apparatus and smart machine | |
| CN105046529A (en) | Mobile advertisement cheating recognition method | |
| CN107566897A (en) | A kind of discrimination method, device and the electronic equipment of video brush amount | |
| CN110335596A (en) | Products Show method, apparatus, equipment and storage medium based on speech recognition | |
| CN108132964A (en) | A kind of collaborative filtering method to be scored based on user item class | |
| CN105023178A (en) | Main body-based electronic commercere commendation method | |
| Ibrahim et al. | Emissions and trade in Southeast and East Asian countries: a panel co-integration analysis | |
| CN105930532A (en) | Method and device of recommending multimedia resources to user | |
| CN111639169A (en) | Man-machine interaction method and device, computer readable storage medium and electronic equipment | |
| CN105653721A (en) | Test effect evaluation method and system | |
| CN106780258A (en) | A kind of method for building up and device of minor crime decision tree | |
| CN110110206A (en) | Relation excavation and recommended method, device, calculating equipment, storage medium between article | |
| Fidrmuc et al. | Meta‐analysis of Chinese business cycle correlation | |
| Lu et al. | On combining visual perception and color structure based image quality assessment | |
| CN107077315A (en) | For select will the voice used with user's communication period system and method | |
| CN109829051B (en) | Method and device for screening similar sentences of database |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160608 |
|
| RJ01 | Rejection of invention patent application after publication |