CN113688861A - Low-dimensional feature small sample multi-classification method and device based on machine learning - Google Patents
Low-dimensional feature small sample multi-classification method and device based on machine learning Download PDFInfo
- Publication number
- CN113688861A CN113688861A CN202110761569.XA CN202110761569A CN113688861A CN 113688861 A CN113688861 A CN 113688861A CN 202110761569 A CN202110761569 A CN 202110761569A CN 113688861 A CN113688861 A CN 113688861A
- Authority
- CN
- China
- Prior art keywords
- feature
- classification
- module
- transformation
- result
- 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
- 238000000034 method Methods 0.000 title claims abstract description 89
- 238000010801 machine learning Methods 0.000 title claims abstract description 33
- 230000009466 transformation Effects 0.000 claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000002790 cross-validation Methods 0.000 claims abstract description 9
- 230000008569 process Effects 0.000 claims abstract description 7
- 238000003379 elimination reaction Methods 0.000 claims description 25
- 230000008030 elimination Effects 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000012706 support-vector machine Methods 0.000 claims description 14
- 238000000513 principal component analysis Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
- G06F18/2113—Selection of the most significant subset of features by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a low-dimensional feature small sample multi-classification method based on machine learning, and relates to the technical field of machine learning, wherein the method comprises the following steps: dividing the sample for multiple times by using a three-fold classification cross validation method to obtain divided samples; carrying out feature selection on the divided samples to obtain a feature selection result; performing feature transformation on the feature selection result to obtain transformed features; and processing the transformed features to obtain a final classification result. By adopting the scheme, the method and the device can analyze and process noisy characteristics, have good robustness on noisy data, have good generalization capability on small sample data, and are suitable for the problems of high difficulty and multiple classifications of high inter-class characteristic overlapping degree, noise characteristics and small training data quantity.
Description
Technical Field
The application relates to the technical field of machine learning, in particular to a low-dimensional feature small sample multi-classification method and device based on machine learning.
Background
With the development of deep learning, various subject fields use deep learning technologies such as neural networks to solve field problems and make a series of breakthrough progresses. However, deep neural network training requires a large amount of data, and the neural network is difficult to function because of the difficulty in data acquisition, high cost, or the absence of a large amount of data.
Disclosure of Invention
The present application is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, a first objective of the present application is to provide a low-dimensional feature small sample multi-classification method based on machine learning, which solves the problem of small sample low-dimensional feature multi-classification with high inter-class feature overlapping degree and noisy features.
The second purpose of the present application is to provide a low-dimensional feature small sample multi-classification device based on machine learning.
A third object of the present application is to propose a non-transitory computer-readable storage medium.
In order to achieve the above object, an embodiment of a first aspect of the present application provides a low-dimensional feature small sample multi-classification method based on machine learning, including: dividing the sample for multiple times by using a three-fold classification cross validation method to obtain divided samples; carrying out feature selection on the divided samples to obtain a feature selection result; performing feature transformation on the feature selection result to obtain transformed features; and processing the transformed features to obtain a final classification result.
Optionally, in an embodiment of the present application, the feature selection specifically includes:
and (3) using a recursive feature elimination method to randomly eliminate one feature each time, selecting a feature elimination method with the best classification result, then recursively removing the next feature until the last feature is left, and counting the elimination methods with the best classification results in all the elimination processes as feature selection results.
Optionally, in an embodiment of the present application, the feature transformation is performed by using a large-space nearest neighbor method, which specifically includes the following steps:
initializing a characteristic transformation matrix by using a Principal Component Analysis (PCA) method;
optimizing a feature transformation matrix by using a large-interval nearest distance method LMNN;
and performing linear transformation on the feature selection result by using the feature transformation matrix to obtain the transformed features.
Optionally, in an embodiment of the present application, a linear kernel support vector machine and a gaussian kernel support vector machine are used to perform classification training on the transformed features respectively to obtain classification probabilities, and the classification probabilities are multiplied and normalized to generate a final classification result.
In order to achieve the above object, a second aspect of the present application provides a low-dimensional feature small sample multi-classification device based on machine learning, including a dividing module, a feature selecting module, a feature transforming module, and a classifier module, wherein,
the dividing module is used for dividing the sample for multiple times by using a three-fold classification cross validation method to obtain the divided sample;
the characteristic selection module is used for carrying out characteristic selection on the divided samples to obtain a characteristic selection result;
the characteristic transformation module is used for carrying out characteristic transformation on the characteristic selection result to obtain transformed characteristics;
and the classifier module is used for processing the transformed features to obtain a final classification result.
Optionally, in an embodiment of the present application, the feature selection module is specifically configured to:
and (3) using a recursive feature elimination method to randomly eliminate one feature each time, selecting a feature elimination method with the best classification result, then recursively removing the next feature until the last feature is left, and counting the elimination methods with the best classification results in all the elimination processes as feature selection results.
Optionally, in an embodiment of the present application, the feature transformation module performs feature transformation by using a large-space nearest neighbor (LMNN) method, specifically including the following steps:
initializing a characteristic transformation matrix by using a Principal Component Analysis (PCA) method;
optimizing a feature transformation matrix by using a large-interval nearest distance method LMNN;
and performing linear transformation on the feature selection result by using the feature transformation matrix to obtain the transformed features.
Optionally, in an embodiment of the present application, the classifier module is specifically configured to:
and (3) using a linear kernel support vector machine and a Gaussian kernel support vector machine to respectively carry out classification training on the transformed features to obtain classification probabilities, multiplying and normalizing the classification probabilities to generate a final classification result.
In order to achieve the above object, a non-transitory computer-readable storage medium is provided in a third aspect of the present application, and instructions in the storage medium are executed by a processor to perform a low-dimensional feature small-sample multi-classification method and apparatus based on machine learning.
The low-dimensional feature small sample multi-classification method based on machine learning, the low-dimensional feature small sample multi-classification device based on machine learning and the non-transitory computer readable storage medium solve the problem of small sample low-dimensional feature multi-classification with high inter-class feature overlapping degree and noise features, and the provided classification method using machine learning technology to perform multi-dimensional feature screening fusion can analyze and process noisy features, is good in robustness to noisy data, has good generalization capability on small sample data, and is suitable for the high-difficulty multi-classification problem with high inter-class feature overlapping degree, noise features and less training data.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of a low-dimensional feature small sample multi-classification method based on machine learning according to an embodiment of the present application;
fig. 2 is a diagram of a raw feature distribution of a low-dimensional feature small sample multi-classification method based on machine learning according to an embodiment of the present application, which is visualized using tSNE;
fig. 3 is a feature distribution diagram after feature selection and feature transformation using tSNE visualization of a low-dimensional feature small sample multi-classification method based on machine learning according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a low-dimensional feature small-sample multi-classification device based on machine learning according to a second embodiment of the present application.
Detailed Description
Reference will now be made in detail to the embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
The method and the device for low-dimensional feature small sample multi-classification based on machine learning of the embodiment of the application are described below with reference to the accompanying drawings.
Fig. 1 is a flowchart of a low-dimensional feature small sample multi-classification method based on machine learning according to an embodiment of the present application.
As shown in fig. 1, the low-dimensional feature small sample multi-classification method based on machine learning includes the following steps:
102, performing feature selection on the divided samples to obtain a feature selection result;
103, performing feature transformation on the feature selection result to obtain transformed features;
and 104, processing the transformed features to obtain a final classification result.
According to the low-dimensional characteristic small sample multi-classification method based on machine learning, a sample is divided for multiple times by using a three-fold classification cross validation method, so that divided samples are obtained; carrying out feature selection on the divided samples to obtain a feature selection result; performing feature transformation on the feature selection result to obtain transformed features; and processing the transformed features to obtain a final classification result. Therefore, the small-sample low-dimensional feature multi-classification problem with high inter-class feature overlapping degree and noise features can be solved, the classification method for multi-dimensional feature screening fusion by using the machine learning technology can analyze and process the noisy features, has good robustness on the noisy data, has good generalization capability on the small-sample data, and is suitable for the high-difficulty multi-classification problems with high inter-class feature overlapping degree, noise features and less training data amount.
Further, in the embodiment of the present application, the feature selection specifically includes:
and (3) using a recursive feature elimination method to randomly eliminate one feature each time, selecting a feature elimination method with the best classification result, then recursively removing the next feature until the last feature is left, and counting the elimination methods with the best classification results in all the elimination processes as feature selection results.
And screening the input low-dimensional features by using a recursive feature elimination method to obtain a feature set with high relevance to classification results. The used low-dimensional input features are generally about 10-dimensional, all the features do not have distinction degree on classification results, and the input features are analyzed and screened, and partial features important to the classification problem are selected to prevent interference caused by irrelevant features.
Further, in the embodiment of the present application, a large-interval closest distance LMNN method is adopted to perform feature transformation, which specifically includes the following steps:
initializing a characteristic transformation matrix by using a Principal Component Analysis (PCA) method;
optimizing a feature transformation matrix by using a large-interval nearest distance method LMNN;
and performing linear transformation on the feature selection result by using the feature transformation matrix to obtain the transformed features.
The feature after feature selection is transformed using a conventional metric learning (metric learning) method with large distance between nearest neighbors (LMNN). A transformation matrix is learned through measurement learning, so that the distance of the same classification sample in a high-dimensional feature space is as close as possible, different classification samples are as far away as possible, the distribution of features is optimized, and inseparable samples in an original feature space are separable as far as possible in a transformed feature space.
Further, in the embodiment of the present application, a linear kernel support vector machine and a gaussian kernel support vector machine are used to perform classification training on the transformed features respectively to obtain classification probabilities, and the classification probabilities are multiplied and normalized to generate a final classification result.
And the linear kernel support vector machine (svm) and the Gaussian kernel support vector machine are used for carrying out weighted voting to carry out multi-classification on the transformed features.
Fig. 2 is a diagram of a raw feature distribution of a low-dimensional feature small sample multi-classification method based on machine learning, which is visualized by tSNE.
Fig. 3 is a feature distribution diagram after feature selection and feature transformation using tSNE visualization of a low-dimensional feature small sample multi-classification method based on machine learning according to an embodiment of the present application.
As shown in fig. 2 and fig. 3, after feature selection and feature transformation are performed on original features by using a low-dimensional feature small sample multi-classification method based on machine learning, the separability of the features is greatly enhanced.
Fig. 4 is a schematic structural diagram of a low-dimensional feature small-sample multi-classification device based on machine learning according to a second embodiment of the present application.
As shown in fig. 4, the low-dimensional feature small sample multi-classification device based on machine learning comprises a division module, a feature selection module, a feature transformation module and a classifier module, wherein,
the dividing module 10 is configured to divide the sample for multiple times by using a three-fold classification cross validation method to obtain divided samples;
the feature selection module 20 is configured to perform feature selection on the divided samples to obtain a feature selection result;
a feature transformation module 30, configured to perform feature transformation on the feature selection result to obtain transformed features;
and the classifier module 40 is used for processing the transformed features to obtain a final classification result.
Further, in this embodiment of the present application, the feature selection module is specifically configured to:
and (3) using a recursive feature elimination method to randomly eliminate one feature each time, selecting a feature elimination method with the best classification result, then recursively removing the next feature until the last feature is left, and counting the elimination methods with the best classification results in all the elimination processes as feature selection results.
Further, in this embodiment of the present application, the feature transformation module performs feature transformation by using a large-space nearest neighbor (LMNN) method, which specifically includes the following steps:
initializing a characteristic transformation matrix by using a Principal Component Analysis (PCA) method;
optimizing a feature transformation matrix by using a large-interval nearest distance method LMNN;
and performing linear transformation on the feature selection result by using the feature transformation matrix to obtain the transformed features.
Further, in this embodiment of the present application, the classifier module is specifically configured to:
and (3) using a linear kernel support vector machine and a Gaussian kernel support vector machine to respectively carry out classification training on the transformed features to obtain classification probabilities, multiplying and normalizing the classification probabilities to generate a final classification result.
Training in an end-to-end mode, and selecting the optimal parameters of the hyper-parameters of different modules by using a grid search (grid search) method. The marker expression quantity features after feature transformation have certain separability, and then the classification is carried out in a mode of fusing a plurality of classifiers, so that the analysis of the plurality of classifiers can be integrated, and the accuracy of classification results is enhanced.
The low-dimensional feature small sample multi-classification device based on machine learning comprises a division module, a feature selection module, a feature transformation module and a classifier module, wherein the division module is used for carrying out multi-time division on a sample by using a three-fold classification cross validation method to obtain divided samples; the characteristic selection module is used for carrying out characteristic selection on the divided samples to obtain a characteristic selection result; the characteristic transformation module is used for carrying out characteristic transformation on the characteristic selection result to obtain transformed characteristics; and the classifier module is used for processing the transformed features to obtain a final classification result. Therefore, the small-sample low-dimensional feature multi-classification problem with high inter-class feature overlapping degree and noise features can be solved, the classification method for multi-dimensional feature screening fusion by using the machine learning technology can analyze and process the noisy features, has good robustness on the noisy data, has good generalization capability on the small-sample data, and is suitable for the high-difficulty multi-classification problems with high inter-class feature overlapping degree, noise features and less training data amount.
In order to achieve the above embodiments, the present application further proposes a non-transitory computer-readable storage medium on which a computer program is stored, the computer program, when executed by a processor, implementing the low-dimensional feature small-sample multi-classification method and apparatus based on machine learning of the above embodiments.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (9)
1. A low-dimensional feature small sample multi-classification method based on machine learning is characterized by comprising the following steps:
dividing the sample for multiple times by using a three-fold classification cross validation method to obtain divided samples;
performing feature selection on the divided samples to obtain a feature selection result;
performing feature transformation on the feature selection result to obtain transformed features;
and processing the transformed features to obtain a final classification result.
2. The method of claim 1, wherein the feature selection comprises the following specific processes:
and (3) using a recursive feature elimination method to randomly eliminate one feature each time, selecting a feature elimination method with the best classification result, then recursively removing the next feature until the last feature is left, and counting the elimination method with the best classification result in all the elimination processes as the feature selection result.
3. The method of claim 1, wherein the feature transformation is performed by using a large-space nearest neighbor LMNN method, which specifically comprises the steps of:
initializing a characteristic transformation matrix by using a Principal Component Analysis (PCA) method;
optimizing the feature transformation matrix by using a large-interval nearest distance method LMNN;
and performing linear transformation on the feature selection result by using the feature transformation matrix to obtain the transformed features.
4. The method of claim 1, wherein the transformed features are classified and trained using a linear kernel support vector machine and a gaussian kernel support vector machine, respectively, to obtain classification probabilities, and the classification probabilities are multiplied and normalized to generate the final classification result.
5. A low-dimensional feature small sample multi-classification device based on machine learning is characterized by comprising a dividing module, a feature selection module, a feature transformation module and a classifier module, wherein,
the dividing module is used for dividing the sample for multiple times by using a three-fold classification cross validation method to obtain divided samples;
the characteristic selection module is used for carrying out characteristic selection on the divided samples to obtain a characteristic selection result;
the feature transformation module is used for carrying out feature transformation on the feature selection result to obtain transformed features;
and the classifier module is used for processing the transformed features to obtain a final classification result.
6. The apparatus of claim 5, wherein the feature selection module is specifically configured to:
and (3) using a recursive feature elimination method to randomly eliminate one feature each time, selecting a feature elimination method with the best classification result, then recursively removing the next feature until the last feature is left, and counting the elimination method with the best classification result in all the elimination processes as the feature selection result.
7. The apparatus of claim 5, wherein the feature transformation module performs the feature transformation by using a large-space nearest neighbor (LMNN) method, and specifically comprises the following steps:
initializing a characteristic transformation matrix by using a Principal Component Analysis (PCA) method;
optimizing the feature transformation matrix by using a large-interval nearest distance method LMNN;
and performing linear transformation on the feature selection result by using the feature transformation matrix to obtain the transformed features.
8. The apparatus of claim 5, wherein the classifier module is specifically configured to:
and respectively carrying out classification training on the transformed features by using a linear kernel support vector machine and a Gaussian kernel support vector machine to obtain classification probabilities, and multiplying and normalizing the classification probabilities to generate the final classification result.
9. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110761569.XA CN113688861A (en) | 2021-07-06 | 2021-07-06 | Low-dimensional feature small sample multi-classification method and device based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110761569.XA CN113688861A (en) | 2021-07-06 | 2021-07-06 | Low-dimensional feature small sample multi-classification method and device based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113688861A true CN113688861A (en) | 2021-11-23 |
Family
ID=78576660
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110761569.XA Pending CN113688861A (en) | 2021-07-06 | 2021-07-06 | Low-dimensional feature small sample multi-classification method and device based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113688861A (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030086593A1 (en) * | 2001-05-31 | 2003-05-08 | Chengjun Liu | Feature based classification |
| CN105117688A (en) * | 2015-07-29 | 2015-12-02 | 重庆电子工程职业学院 | Face identification method based on texture feature fusion and SVM |
| US20160379138A1 (en) * | 2013-12-12 | 2016-12-29 | Indian Institute Of Technology Delhi | Classifying test data based on a maximum margin classifier |
| CN109165672A (en) * | 2018-07-16 | 2019-01-08 | 华南理工大学 | A kind of Ensemble classifier method based on incremental learning |
| CN109711461A (en) * | 2018-12-25 | 2019-05-03 | 中国人民解放军战略支援部队航天工程大学 | Transfer learning picture classification method and its device based on principal component analysis |
| CN109961093A (en) * | 2019-03-07 | 2019-07-02 | 北京工业大学 | A kind of image classification method based on many intelligence integrated studies |
| CN110705722A (en) * | 2019-09-12 | 2020-01-17 | 华中科技大学 | Diagnostic model for industrial equipment fault diagnosis and construction method and application thereof |
| CN110956187A (en) * | 2019-11-28 | 2020-04-03 | 中国农业科学院农业信息研究所 | Unmanned aerial vehicle image plant canopy information extraction method based on ensemble learning |
| CN111008637A (en) * | 2018-10-08 | 2020-04-14 | 北京京东尚科信息技术有限公司 | Image classification method and system |
| CN111009321A (en) * | 2019-08-14 | 2020-04-14 | 电子科技大学 | Application method of machine learning classification model in juvenile autism auxiliary diagnosis |
| CN111753987A (en) * | 2020-07-08 | 2020-10-09 | 深延科技(北京)有限公司 | Method and device for generating machine learning model |
| CN112070125A (en) * | 2020-08-19 | 2020-12-11 | 西安理工大学 | Prediction method of unbalanced data set based on isolated forest learning |
| CN112102899A (en) * | 2020-09-15 | 2020-12-18 | 北京晶派科技有限公司 | Construction method of molecular prediction model and computing equipment |
-
2021
- 2021-07-06 CN CN202110761569.XA patent/CN113688861A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030086593A1 (en) * | 2001-05-31 | 2003-05-08 | Chengjun Liu | Feature based classification |
| US20160379138A1 (en) * | 2013-12-12 | 2016-12-29 | Indian Institute Of Technology Delhi | Classifying test data based on a maximum margin classifier |
| CN105117688A (en) * | 2015-07-29 | 2015-12-02 | 重庆电子工程职业学院 | Face identification method based on texture feature fusion and SVM |
| CN109165672A (en) * | 2018-07-16 | 2019-01-08 | 华南理工大学 | A kind of Ensemble classifier method based on incremental learning |
| CN111008637A (en) * | 2018-10-08 | 2020-04-14 | 北京京东尚科信息技术有限公司 | Image classification method and system |
| CN109711461A (en) * | 2018-12-25 | 2019-05-03 | 中国人民解放军战略支援部队航天工程大学 | Transfer learning picture classification method and its device based on principal component analysis |
| CN109961093A (en) * | 2019-03-07 | 2019-07-02 | 北京工业大学 | A kind of image classification method based on many intelligence integrated studies |
| CN111009321A (en) * | 2019-08-14 | 2020-04-14 | 电子科技大学 | Application method of machine learning classification model in juvenile autism auxiliary diagnosis |
| CN110705722A (en) * | 2019-09-12 | 2020-01-17 | 华中科技大学 | Diagnostic model for industrial equipment fault diagnosis and construction method and application thereof |
| CN110956187A (en) * | 2019-11-28 | 2020-04-03 | 中国农业科学院农业信息研究所 | Unmanned aerial vehicle image plant canopy information extraction method based on ensemble learning |
| CN111753987A (en) * | 2020-07-08 | 2020-10-09 | 深延科技(北京)有限公司 | Method and device for generating machine learning model |
| CN112070125A (en) * | 2020-08-19 | 2020-12-11 | 西安理工大学 | Prediction method of unbalanced data set based on isolated forest learning |
| CN112102899A (en) * | 2020-09-15 | 2020-12-18 | 北京晶派科技有限公司 | Construction method of molecular prediction model and computing equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8682813B2 (en) | Sample class prediction method, prediction program, and prediction apparatus | |
| CN111134664B (en) | Epileptic discharge identification method and system based on capsule network and storage medium | |
| US11562294B2 (en) | Apparatus and method for analyzing time-series data based on machine learning | |
| JP2018116364A (en) | Dictionary generation device, evaluation device, dictionary generation method, evaluation method, and program | |
| Sani et al. | Redefining selection of features and classification algorithms for room occupancy detection | |
| CN116453438B (en) | Display screen parameter detection method, device, equipment and storage medium | |
| CN112905412A (en) | Method and device for detecting abnormity of key performance index data | |
| CN108470194B (en) | Feature screening method and device | |
| Gupta et al. | Handling class overlapping to detect noisy instances in classification | |
| Bergmeir | Enhanced machine learning and data mining methods for Analysing large hybrid electric vehicle fleets based on load spectrum data | |
| CN109255389B (en) | Equipment evaluation method, device, equipment and readable storage medium | |
| Oliveira-Santos et al. | Combining classifiers with decision templates for automatic fault diagnosis of electrical submersible pumps | |
| CN103221965A (en) | Systems and methods for robust pattern classification | |
| CN113688861A (en) | Low-dimensional feature small sample multi-classification method and device based on machine learning | |
| JP7206892B2 (en) | Image inspection device, learning method for image inspection, and image inspection program | |
| CN107067034A (en) | A kind of method and system of quick identification ir data classification | |
| Mur et al. | Unsupervised event detection and classification of multichannel signals | |
| Al-Qadasi et al. | DeepAbstraction++: Enhancing Test Prioritization Performance via Combined Parameterized Boxes | |
| CN113139332A (en) | Automatic model construction method, device and equipment | |
| CN112434624A (en) | Structural health monitoring data distortion detection method, system, device and storage medium | |
| Vareille et al. | Evaluating Explanation Methods of Multivariate Time Series Classification through Causal Lenses | |
| CN117313899B (en) | Method, apparatus and medium for data processing | |
| CN117171141B (en) | Data model modeling method based on relational graph | |
| CN117313900B (en) | Method, apparatus and medium for data processing | |
| JP2018088079A (en) | Learning device and learning method |
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 |