CN115859059A - Repeatable labeling method, system and device for fuzzy information - Google Patents
Repeatable labeling method, system and device for fuzzy information Download PDFInfo
- Publication number
- CN115859059A CN115859059A CN202211026598.2A CN202211026598A CN115859059A CN 115859059 A CN115859059 A CN 115859059A CN 202211026598 A CN202211026598 A CN 202211026598A CN 115859059 A CN115859059 A CN 115859059A
- Authority
- CN
- China
- Prior art keywords
- fuzzy membership
- labeling
- fuzzy
- tensor
- sample
- 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.)
- Granted
Links
- 238000002372 labelling Methods 0.000 title claims abstract description 113
- 238000000034 method Methods 0.000 claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims abstract description 22
- 238000012937 correction Methods 0.000 claims abstract description 14
- 230000010354 integration Effects 0.000 claims abstract description 4
- 238000005315 distribution function Methods 0.000 claims description 17
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000007812 deficiency Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000002904 solvent Substances 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a system and a device for repeatedly marking fuzzy information, wherein the method comprises the following steps: acquiring an annotation task and determining an annotation mode, a sample to be annotated and an annotation main body; based on the labeling mode, labeling the to-be-labeled sample by the labeling main body to obtain corresponding fuzzy membership degrees, and integrating the fuzzy membership degrees corresponding to the labeling main bodies to obtain an initial fuzzy membership degree tensor; and estimating the distribution of the fuzzy membership degree and the preference of a labeling main body, and correcting the initial fuzzy membership degree tensor to obtain a final fuzzy membership degree matrix. The system comprises: the device comprises an acquisition module, a marking module, an integration module and a correction module. The apparatus includes a memory and a processor for performing the repeatable labeling method of the ambiguous information described above. By using the method and the device, the fuzzy marking data can be corrected from the repeated marking of the fuzzy information, and the influence of missing data on the marking quality is reduced. The invention can be widely applied to the field of data information labeling.
Description
Technical Field
The present invention relates to the field of information labeling, and in particular, to a method, a system, and an apparatus for repeatable labeling of fuzzy information.
Background
At present, fuzzy information crowdsourcing annotation work generally involves the following problems: 1. a single sample can be repeatedly marked with fuzzy information by a plurality of marking personnel, and the marking results of the sample cannot be unified; 2. the number of related labeling personnel of the samples to be labeled in the same batch is large, and the problems of personal labeling preference difference and labeling knowledge difference exist, so that the labeling standards among the samples are difficult to be consistent; based on the above reasons, a large amount of repeatable labeling data of heterogeneous unconstrained fuzzy information with different preference information and missing information is generated, and a fuzzy data processing method capable of correcting fuzzy labeling from repeated fuzzy information labeling and reducing the influence of partial data missing in the fuzzy information is urgently needed.
Disclosure of Invention
In order to solve the above technical problems, an object of the present invention is to provide a method, a system, and an apparatus for repeatable labeling of fuzzy information, which can correct fuzzy labeled data from repeated labeling of fuzzy information, and reduce the influence of missing data on labeling quality.
The first technical scheme adopted by the invention is as follows: a repeatable labeling method of fuzzy information comprises the following steps:
acquiring an annotation task and determining an annotation mode, a sample to be annotated and an annotation subject;
based on the labeling mode, labeling the to-be-labeled sample by the labeling main body to obtain the corresponding fuzzy membership degree;
integrating fuzzy membership degrees corresponding to the plurality of marking main bodies to obtain an initial fuzzy membership degree tensor;
and estimating fuzzy membership distribution and marking subject preference, and correcting the initial fuzzy membership tensor to obtain a final fuzzy membership matrix.
Further, the sample to be labeled is a repeatable labeling sample, the labeling main bodies are labeling experts with different preferences, and the labeling constraint condition of the fuzzy membership is an unconstrained labeling condition.
Further, the step of integrating the fuzzy membership corresponding to the plurality of labeling subjects to obtain an initial fuzzy membership tensor specifically includes:
constructing fuzzy membership degree matrixes of different labeling experts according to fuzzy membership degrees corresponding to the labeling experts;
and summarizing the fuzzy membership degree matrixes of all the labeled experts to form an initial fuzzy membership degree tensor.
Further, the step of predicting the distribution of the fuzzy membership degree and labeling the preference of the subject and correcting the initial fuzzy membership degree tensor to obtain a final fuzzy membership degree matrix specifically comprises the following steps:
based on a large number law, estimating a sample to be labeled and a fuzzy membership obeying normal distribution function to obtain a fuzzy membership distribution function;
acquiring the maximum value and the minimum value of the fuzzy membership according to the initial fuzzy membership tensor;
estimating a preference coefficient of the labeling expert according to the maximum value and the minimum value of the fuzzy membership degree to obtain an expert preference coefficient pre-estimated value;
and correcting the fuzzy membership tensor according to the fuzzy membership distribution function and the expert preference coefficient pre-estimation value to obtain a final fuzzy membership matrix.
Further, the fuzzy membership follows a normal distribution, and the formula is as follows:
further, the step of correcting the fuzzy membership tensor according to the fuzzy membership distribution function and the expert preference coefficient pre-estimation value to obtain a final fuzzy membership matrix specifically includes:
determining a distribution interval according to a fuzzy membership distribution function;
calculating the average value of the expert preference coefficient estimated values;
combining the average value of the expert preference coefficient pre-estimated value with the distribution interval to obtain correction data;
and correcting the fuzzy membership tensor according to the correction data to obtain a final fuzzy membership matrix.
The second technical scheme adopted by the invention is as follows: a repeatable annotation system of fuzzy information comprising:
the acquisition module is used for acquiring the labeling task and determining a labeling mode, a sample to be labeled and a labeling main body;
the labeling module is used for labeling the to-be-labeled sample by the labeling main body based on the labeling mode to obtain the corresponding fuzzy membership degree;
the integration module is used for integrating the fuzzy membership degrees corresponding to the plurality of marking main bodies to obtain an initial fuzzy membership degree tensor;
and the correction module is used for predicting the fuzzy membership degree distribution and the labeling main body preference and correcting the initial fuzzy membership degree tensor to obtain a final fuzzy membership degree matrix.
The third technical scheme adopted by the invention is as follows: a repeatable labeling apparatus for obscuring information, comprising:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to implement a repeatable method of annotating fuzzy information as described above.
The method, the system and the device have the advantages that: the invention designs a fuzzy information heterogeneous unconstrained repeatable labeling method, establishes a fuzzy membership grade labeling tensor, estimates fuzzy membership grade distribution according to fuzzy information labeling data, estimates preference information of different labeling main bodies, and corrects a fuzzy label from a heterogeneous unconstrained repeatable label, thereby achieving the aims of correcting the fuzzy label from repeated fuzzy information labeling and reducing the influence of missing data on labeling quality.
Drawings
FIG. 1 is a flow chart illustrating steps of a method for repeatedly labeling fuzzy information according to the present invention;
FIG. 2 is a block diagram of a repeatable annotation system for fuzzy information according to the present invention;
fig. 3 is a schematic diagram of a labeling process in an application scenario according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of a correction process in an application scenario according to an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the specific embodiments. The step numbers in the following embodiments are provided only for convenience of illustration, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.
Referring to fig. 1, 3 and 4, the present invention provides a repeatable labeling method of fuzzy information, comprising the steps of:
s1, obtaining an annotation task and determining an annotation mode, a sample to be annotated and an annotation subject;
specifically, the to-be-labeled sample is a repeatable labeling sample, the labeling main bodies are labeling experts with different preferences, and the labeling constraint conditions of the to-be-fuzzy membership degree are unconstrained labeling conditions. The expert preferences are represented by preference coefficients p, p r (R =1,2,3.. R) represents the R-th expert preference, the higher the expert preference coefficient, indicating that the higher the expert labeled fuzzy membership when the sample is in large relation to the labeled pattern, and indicating that the lower the expert labeled fuzzy membership when the sample is in small relation to the labeled pattern.
Annotation service demander typically provides an annotation standard (annotation schema)
And giving a labeling expert for labeling the sample to be labeled.
X＝{x 1 ,x 2 ,...,x n Is the set of samples to be labeled, sample x j Represents the j-th sample in X,
represents a sample x j (j =1, 2.. N.) in relation to the annotation pattern->
Fuzzy membership of (c). If->
Then sample x is weighed j Regarding the marking mode->
Is fuzzy membership degree->
And satisfying strong constraints (the corresponding labeling constraint condition is called as a strong constraint labeling condition), otherwise, calling satisfying weak constraints (the corresponding labeling constraint condition is called as a weak constraint labeling condition). If fuzzy membership degree>
If the strong constraint or the weak constraint is not required to be satisfied, the labeling constraint condition is called to satisfy the unconstrained condition (the corresponding labeling constraint condition is the unconstrained labeling condition).
When the r-th expert marks a part of the sample, the sample x j (j =1, 2.. N.) about annotation patterns
Is fuzzy membership degree->
Fuzzy membership degree matrix forming the formula>
Wherein->
Sample x representing the r-th expert label j With respect to marking mode>
Is fuzzy membership degree->
Contains the r-th expert preference coefficient rho r And mark mode>
The information of (a).
Summarizing fuzzy membership matrix of R experts
An initial fuzzy membership tensor @, constituting dimensions c x n x R>
S2, based on the labeling mode, labeling the to-be-labeled sample by the labeling main body to obtain a corresponding fuzzy membership degree;
s3, integrating fuzzy membership degrees corresponding to the plurality of marking main bodies to obtain an initial fuzzy membership degree tensor;
s3.1, constructing fuzzy membership degree matrixes of different labeling experts according to fuzzy membership degrees corresponding to the labeling experts;
and S3.2, summarizing the fuzzy membership matrixes of all the labeled experts to form an initial fuzzy membership tensor.
Specifically, in the labeling stage, the r-th expert empirically judges the sample x j And annotation schema
Is marked with the relation of (1), sample x j With respect to marking mode>
Is blurredMembership degree->
The r-th expert labeled sample x j About annotation patterns
The fuzzy membership of can be formed into vectors
After the r-th expert labels a partial sample,
fuzzy membership matrix capable of forming the label of the r-th expert>
Summarizing fuzzy membership degree matrix of R experts
Composing an initial fuzzy membership tensor>
And S4, pre-estimating fuzzy membership distribution and marking subject preference, and correcting the initial fuzzy membership tensor to obtain a final fuzzy membership matrix.
Obtaining fuzzy membership degree matrixes of R experts through initial fuzzy membership degree tensor labeling process
There are two types of deficiency values: first kind of deficiency valueThe method comprises the steps that samples which are not labeled by experts are generated, and fuzzy membership degrees of the samples relative to all labeled modes are missing values; the second missing value is generated by human factors or hardware loss, and the fuzzy membership of the sample relative to the partial labeling mode is the missing value. The first missing value is referred to as a complete missing value and the second missing value is referred to as a partial missing value. The following formula shows sample x 2 Is a complete missing value, and sample x j Regarding the marking mode->
In the case of partial absence of fuzzy membership value(s), in the case of a fuzzy membership value(s) in the evaluation unit>
In the case of (3), NA is the deletion value:
the fuzzy membership degree correction process comprises three stages: stage one, estimating fuzzy membership degree distribution of each sample; stage two pre-estimating expert preference; and step three, correcting the fuzzy membership by utilizing fuzzy membership distribution and expert preference pre-evaluation value.
S4.1, based on a majority law, estimating a sample to be labeled and a fuzzy membership obeying normal distribution function to obtain a fuzzy membership distribution function;
sample x after initial fuzzy membership tensor labeling process j R experts in sample x j About annotation patterns
Is based on a fuzzy membership degree of->
Based on the law of large numbers, in combination with a suitable selection of a number or a number of combinations>
Obey a normal distribution. Sample x j Regarding the marking mode->
Is normally distributed according to the formula
Wherein:
wherein R is r Samples x labeled for R experts j About annotation patterns
Degree of fuzzy membership of
In satisfies >>
The number of (2), R r R is less than or equal to R. Initial fuzzy membership tensor
Fuzzy membership in (4)>
Obeying a normal distribution function
Thus, the initial fuzzy membership tensor pick>
Obey is selected by>
Forming a normal distribution function matrix:
s4.2, acquiring the maximum value and the minimum value of the fuzzy membership degree according to the initial fuzzy membership degree tensor;
s4.3, pre-estimating the preference coefficient of the labeled expert according to the maximum value and the minimum value of the fuzzy membership degree to obtain an expert preference coefficient pre-estimated value;
in particular, the amount of the solvent to be used,
the kth sample, marked for the r-th expert>
And
are respectively a sample->
Regarding the marking mode->
Fuzzy degree of membership of
Maximum and minimum values of (a).
K (r) is the number of samples labeled by the r-th expert, the preference p of the r-th expert r And fuzzy degree of membership
Is greater than or equal to>
And a minimum value->
And (4) correlating. The preference coefficient prediction for the r expert is obtained in accordance with the formula>
And c is the number of the labeled patterns.
Therefore, if expert
All labeled sample x j ，/>
For labeling sample x j H (x) of j ) H expert of the experts, sample x j Underlying average expert preference coefficient>
Is composed of
And S4.4, correcting the fuzzy membership tensor according to the fuzzy membership distribution function and the expert preference coefficient pre-estimation value to obtain a final fuzzy membership matrix.
S4.4.1, determining a distribution interval according to a fuzzy membership distribution function;
s4.4.2, calculating the average value of the estimated values of the expert preference coefficients;
s4.4.3, combining the average value of the expert preference coefficient estimated values with the distribution interval to obtain correction data;
and S4.4.4, correcting the fuzzy membership tensor according to the correction data to obtain a final fuzzy membership matrix.
The vast majority of fuzzy membership
The numerical values are all in 3 sigma of the fuzzy membership distribution function ij Interval (mu) ij -3σ ij ,μ ij +3σ ij ) Inner, sample x j Lower average expert preference factor->
Binding Interval (. Mu.) ij -3σ ij ,μ ij +3σ ij ) Designing a repeatable fuzzy membership label correction method to obtain a final fuzzy membership matrix->
Where θ is the threshold, sample x j In that
In respect of the marking mode>
Is based on the final fuzzy membership degree->
Is composed of
In particular, if R experts are at sample x j If the fuzzy membership degree of all the labeled patterns is a complete missing value, then order
If R experts are in sample x j Regarding a mode->
If the fuzzy membership degrees are partial missing values, the command is given>
Compared with the existing labeling method, the method is more suitable for the adoption of large-scale fuzzy information labeling data, more effectively treats the deficiency problem and solves the repeatable fuzzy information label correction problem.
As shown in fig. 2, a repeatable annotation system for fuzzy information includes:
the acquisition module is used for acquiring the labeling task and determining a labeling mode, a sample to be labeled and a labeling main body;
the labeling module is used for labeling the to-be-labeled sample by the labeling main body based on the labeling mode to obtain the corresponding fuzzy membership degree;
the integration module is used for integrating the fuzzy membership degrees corresponding to the plurality of marking main bodies to obtain an initial fuzzy membership degree tensor;
and the correction module is used for predicting the fuzzy membership distribution and the labeling main body preference and correcting the initial fuzzy membership tensor to obtain a final fuzzy membership matrix.
The contents in the above method embodiments are all applicable to the present system embodiment, the functions specifically implemented by the present system embodiment are the same as those in the above method embodiment, and the beneficial effects achieved by the present system embodiment are also the same as those achieved by the above method embodiment.
A repeatable labeling device of fuzzy information comprises:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to implement a repeatable method of annotating fuzzy information as described above.
The contents in the above method embodiments are all applicable to the present apparatus embodiment, the functions specifically implemented by the present apparatus embodiment are the same as those in the above method embodiments, and the advantageous effects achieved by the present apparatus embodiment are also the same as those achieved by the above method embodiments.
A storage medium having stored therein instructions executable by a processor, the storage medium comprising: the processor-executable instructions, when executed by a processor, are for implementing a repeatable labeling method of ambiguous information as described above.
The contents in the foregoing method embodiments are all applicable to this storage medium embodiment, the functions specifically implemented by this storage medium embodiment are the same as those in the foregoing method embodiments, and the beneficial effects achieved by this storage medium embodiment are also the same as those achieved by the foregoing method embodiments.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A repeatable labeling method of fuzzy information is characterized by comprising the following steps:
acquiring an annotation task and determining an annotation mode, a sample to be annotated and an annotation main body;
based on the labeling mode, labeling the to-be-labeled sample by the labeling main body to obtain the corresponding fuzzy membership degree;
integrating fuzzy membership degrees corresponding to the plurality of marking main bodies to obtain an initial fuzzy membership degree tensor;
and estimating fuzzy membership distribution and marking subject preference, and correcting the initial fuzzy membership tensor to obtain a final fuzzy membership matrix.
2. The method for repeatable labeling of fuzzy information according to claim 1, wherein the samples to be labeled are repeatable labeling samples, the labeling subjects are labeling experts with different preferences, and the labeling constraint condition of the fuzzy membership is an unconstrained labeling condition.
3. The method of claim 2, wherein the step of integrating the fuzzy membership corresponding to the labeled subjects to obtain an initial fuzzy membership tensor comprises:
constructing fuzzy membership degree matrixes of different labeling experts according to fuzzy membership degrees corresponding to the labeling experts;
and summarizing the fuzzy membership degree matrixes of all the labeling experts to form an initial fuzzy membership degree tensor.
4. The method as claimed in claim 3, wherein the step of estimating the distribution of fuzzy membership and the preference of the labeled subject and correcting the initial fuzzy membership tensor to obtain the final fuzzy membership matrix comprises:
based on a large number law, estimating a sample to be labeled and a fuzzy membership obeying normal distribution function to obtain a fuzzy membership distribution function;
acquiring the maximum value and the minimum value of the fuzzy membership according to the initial fuzzy membership tensor;
estimating and labeling the preference coefficient of the expert according to the maximum value and the minimum value of the fuzzy membership degree to obtain an expert preference coefficient estimated value;
and correcting the fuzzy membership tensor according to the fuzzy membership distribution function and the expert preference coefficient pre-evaluation value to obtain a final fuzzy membership matrix.
5. The method of claim 4, wherein the fuzzy membership follows a normal distribution, and the formula is as follows:
6. the method as claimed in claim 5, wherein the step of calibrating the fuzzy membership tensor according to the fuzzy membership distribution function and the expert preference coefficient pre-estimated value to obtain the final fuzzy membership matrix specifically comprises:
determining a distribution interval according to a fuzzy membership distribution function;
calculating the average value of the expert preference coefficient estimated values;
combining the average value of the expert preference coefficient pre-estimated value with the distribution interval to obtain correction data;
and correcting the fuzzy membership tensor according to the correction data to obtain a final fuzzy membership matrix.
7. A repeatable annotation system for fuzzy information, comprising:
the acquisition module is used for acquiring the labeling task and determining a labeling mode, a sample to be labeled and a labeling main body;
the labeling module is used for labeling the to-be-labeled sample by the labeling main body based on the labeling mode to obtain the corresponding fuzzy membership degree;
the integration module is used for integrating the fuzzy membership degrees corresponding to the plurality of marking main bodies to obtain an initial fuzzy membership degree tensor;
and the correction module is used for predicting the fuzzy membership distribution and the labeling main body preference and correcting the initial fuzzy membership tensor to obtain a final fuzzy membership matrix.
8. A repeatable labeling apparatus for obscuring information, comprising:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to implement a method of repeatable labelling of obscured information according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211026598.2A CN115859059B (en) | 2022-08-25 | 2022-08-25 | Repeatable labeling method, system and device for fuzzy information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211026598.2A CN115859059B (en) | 2022-08-25 | 2022-08-25 | Repeatable labeling method, system and device for fuzzy information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115859059A true CN115859059A (en) | 2023-03-28 |
| CN115859059B CN115859059B (en) | 2024-03-22 |
Family
ID=85660630
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211026598.2A Active CN115859059B (en) | 2022-08-25 | 2022-08-25 | Repeatable labeling method, system and device for fuzzy information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115859059B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070112695A1 (en) * | 2004-12-30 | 2007-05-17 | Yan Wang | Hierarchical fuzzy neural network classification |
| CN104199852A (en) * | 2014-08-12 | 2014-12-10 | 上海交通大学 | Label propagation community structure mining method based on node membership degree |
| CN106600605A (en) * | 2016-12-14 | 2017-04-26 | 陕西科技大学 | Unsupervised fast image segmentation algorithm |
| CN109389608A (en) * | 2018-10-19 | 2019-02-26 | 山东大学 | There is the fuzzy clustering image partition method of noise immunity using plane as cluster centre |
| CN109495296A (en) * | 2018-11-02 | 2019-03-19 | 国网四川省电力公司电力科学研究院 | Intelligent substation communication network state evaluation method based on clustering and neural network |
| CN112560921A (en) * | 2020-12-10 | 2021-03-26 | 百维金科(上海)信息科技有限公司 | Internet financial platform application fraud detection method based on fuzzy C-mean |
| CN113177358A (en) * | 2021-04-30 | 2021-07-27 | 燕山大学 | Soft measurement method for cement quality based on fuzzy fine-grained feature extraction |
| CN113688258A (en) * | 2021-08-20 | 2021-11-23 | 广东工业大学 | Information recommendation method and system based on flexible multidimensional clustering |
| CN114266321A (en) * | 2021-12-31 | 2022-04-01 | 广东泰迪智能科技股份有限公司 | Weak supervision fuzzy clustering algorithm based on unconstrained prior information mode |
| CN114330574A (en) * | 2021-12-31 | 2022-04-12 | 广东泰迪智能科技股份有限公司 | Fuzzy labeling method for pattern recognition |
| CN115392376A (en) * | 2022-08-25 | 2022-11-25 | 广东工业大学 | Method, system and device for labeling heterogeneous fuzzy membership degree matrix |
| US20220398478A1 (en) * | 2020-12-15 | 2022-12-15 | Hainan University | Remote sensing image feature discretization method based on rough-fuzzy model |
-
2022
- 2022-08-25 CN CN202211026598.2A patent/CN115859059B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070112695A1 (en) * | 2004-12-30 | 2007-05-17 | Yan Wang | Hierarchical fuzzy neural network classification |
| CN104199852A (en) * | 2014-08-12 | 2014-12-10 | 上海交通大学 | Label propagation community structure mining method based on node membership degree |
| CN106600605A (en) * | 2016-12-14 | 2017-04-26 | 陕西科技大学 | Unsupervised fast image segmentation algorithm |
| CN109389608A (en) * | 2018-10-19 | 2019-02-26 | 山东大学 | There is the fuzzy clustering image partition method of noise immunity using plane as cluster centre |
| CN109495296A (en) * | 2018-11-02 | 2019-03-19 | 国网四川省电力公司电力科学研究院 | Intelligent substation communication network state evaluation method based on clustering and neural network |
| CN112560921A (en) * | 2020-12-10 | 2021-03-26 | 百维金科(上海)信息科技有限公司 | Internet financial platform application fraud detection method based on fuzzy C-mean |
| US20220398478A1 (en) * | 2020-12-15 | 2022-12-15 | Hainan University | Remote sensing image feature discretization method based on rough-fuzzy model |
| CN113177358A (en) * | 2021-04-30 | 2021-07-27 | 燕山大学 | Soft measurement method for cement quality based on fuzzy fine-grained feature extraction |
| CN113688258A (en) * | 2021-08-20 | 2021-11-23 | 广东工业大学 | Information recommendation method and system based on flexible multidimensional clustering |
| CN114266321A (en) * | 2021-12-31 | 2022-04-01 | 广东泰迪智能科技股份有限公司 | Weak supervision fuzzy clustering algorithm based on unconstrained prior information mode |
| CN114330574A (en) * | 2021-12-31 | 2022-04-12 | 广东泰迪智能科技股份有限公司 | Fuzzy labeling method for pattern recognition |
| CN115392376A (en) * | 2022-08-25 | 2022-11-25 | 广东工业大学 | Method, system and device for labeling heterogeneous fuzzy membership degree matrix |
Non-Patent Citations (2)
| Title |
|---|
| KONGSOROT Y等: "Kernel extreme learning machine based on fuzzy set theory for multi-label classification", 《INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS》, pages 979 - 989 * |
| 徐圣兵;林上钧;钟国祥;: "基于交叉熵测度的成对约束模糊核聚类算法", 应用科技, vol. 47, no. 01, pages 80 - 87 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115859059B (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7519446B2 (en) | Manufacture condition setting system, manufacture condition setting method, control program, and computer-readable record medium recording control program therein | |
| CN112232407B (en) | Neural network model training method and device for pathological image samples | |
| CN112232426B (en) | Training method, device and equipment of target detection model and readable storage medium | |
| CN110110726A (en) | The recognition methods of power equipment nameplate, device, computer equipment and storage medium | |
| CN111461101B (en) | Method, device, equipment and storage medium for identifying work clothes mark | |
| CN113221282B (en) | Aero-engine service life prediction method based on Bayesian residual convolutional network | |
| CN111898613B (en) | Semi-supervised semantic segmentation model training method, recognition method and device | |
| CN111199186A (en) | Image quality scoring model training method, device, equipment and storage medium | |
| CN116596582A (en) | Marketing information prediction method and device based on big data | |
| CN115392376B (en) | Heterogeneous fuzzy membership matrix labeling method, system and device | |
| CN113535481B (en) | Data backtracking method and device and non-volatile computer readable storage medium | |
| CN115859059A (en) | Repeatable labeling method, system and device for fuzzy information | |
| US20020181767A1 (en) | Classification of photos with sepia tones | |
| CN117455037A (en) | Multi-mode case knowledge graph-based production business process remaining time prediction method | |
| US20030022404A1 (en) | System processing time computation method, system processing time computation device, and recording medium with system processing time computation program recorded thereon | |
| CN112950621B (en) | Image processing method, device, equipment and medium | |
| CN112328951B (en) | Processing method of experimental data of analysis sample | |
| CN113240076A (en) | Clock error abnormal data detection method based on clustering and neural network algorithm | |
| CN113313179B (en) | Noise image classification method based on l2p norm robust least square method | |
| CN117151227B (en) | Reasoning method and device for semiconductor detection result | |
| US20240046623A1 (en) | Image selection method applied to model retraining | |
| CN112017194B (en) | Image segmentation method, device, equipment and storage medium | |
| CN115357769B (en) | Government affair data labeling system and method | |
| CN116703622B (en) | Vehicle damage identification method and system | |
| US20210325837A1 (en) | Information processing apparatus, information processing method and computer program 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |