CN108664992A - A kind of sorting technique and device based on genetic optimization and core extreme learning machine - Google Patents
A kind of sorting technique and device based on genetic optimization and core extreme learning machine Download PDFInfo
- Publication number
- CN108664992A CN108664992A CN201810295090.XA CN201810295090A CN108664992A CN 108664992 A CN108664992 A CN 108664992A CN 201810295090 A CN201810295090 A CN 201810295090A CN 108664992 A CN108664992 A CN 108664992A
- Authority
- CN
- China
- Prior art keywords
- picture
- learning machine
- extreme learning
- functional parameter
- population
- 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
Classifications
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Abstract
The sorting technique based on genetic optimization and core extreme learning machine that the invention discloses a kind of, including:Picture in sample graph valut is input in recurrent least square method core extreme learning machine;Its nicety of grading classified to the picture in sample graph valut under multigroup different regular coefficient and kernel functional parameter is obtained, classification results include that the user of preset quantity likes grade;Regular coefficient is optimized with kernel functional parameter based on genetic algorithm, obtains optimal regular coefficient and kernel functional parameter;The value of optimal regular coefficient and kernel functional parameter is input in recurrent least square method core extreme learning machine, the picture in picture library to be sorted is classified;Grade is liked to recommend from high to low to user according to corresponding user in picture.Sorting parameter in this method in recurrent least square method core extreme learning machine be based on genetic algorithm optimization after obtained optimal classification parameter, nicety of grading is improved, and then improves picture and recommend accuracy.
Description
Technical field
The present invention relates to data classification technology fields, and genetic optimization and the core limit are based on more specifically to one kind
The sorting technique of habit machine, further relating to a kind of sorter, equipment and computer based on genetic optimization and core extreme learning machine can
Read storage medium.
Background technology
Picture recommends application or picture to recommend webpage that can recommend picture, wherein data to user according to the preference of user
Classification is most important for picture is able to accurately recommend.Current existing data classification method has neural network, supports
Vector machine method, Decision-Tree Method, Bayesian network method, recurrent least square method core extreme learning machine etc., wherein recurrence are most
Small square law core extreme learning machine is a kind of common data taxonomic methods, but user needs when using recurrent least square method
Sorting parameter is chosen according to test result after repeated multiple times experiment, this way is easy when data volume is prodigious when
Cause the selection that result causes sorting parameter because of the subjective judgement of user unreliable, causes nicety of grading relatively low, and then cause
Picture recommends accuracy to reduce, that is, recommends the picture of user and reduced by the favorite probability of user.
It is current those skilled in the art in conclusion how to provide a kind of scheme that can be improved picture and recommend accuracy
Urgent problem to be solved.
Invention content
The object of the present invention is to provide a kind of sorting technique, device, equipment based on genetic optimization and core extreme learning machine
And computer readable storage medium, picture can be improved and recommend accuracy.
To achieve the goals above, the present invention provides the following technical solutions:
A kind of sorting technique based on genetic optimization and core extreme learning machine, including:
Picture in sample graph valut is input in recurrent least square method core extreme learning machine;
The recurrent least square method core extreme learning machine is obtained in multigroup different regular coefficient and kernel functional parameter
Under the nicety of grading classified to the picture in the sample graph valut, wherein one group of normalization coefficient and kernel functional parameter
A corresponding nicety of grading, classification of the recurrent least square method core extreme learning machine to the picture in the sample graph valut
As a result the user for including preset quantity likes grade;
The regular coefficient is optimized with kernel functional parameter based on genetic algorithm, obtains optimal regular coefficient
With kernel functional parameter, wherein any group of normalization coefficient nicety of grading corresponding with kernel functional parameter is suitable in genetic algorithm
Answer angle value;
The value of the optimal regular coefficient and kernel functional parameter is input to the recurrent least square method core limit
In learning machine, the picture in picture library to be sorted is classified;
Grade is liked to recommend from high to low to user according to corresponding user in picture.
Preferably, described that the regular coefficient is optimized with kernel functional parameter based on genetic algorithm, it obtains optimal
Regular coefficient include with kernel functional parameter:
It is all a using multigroup different regular coefficient and kernel functional parameter as the individual of population in genetic algorithm
Body forms the first population;
The second population with default individual amount is randomly generated from first population;
The individual that fitness value is less than preset first fitness value is rejected from second population;
Calculated crosswise is carried out to first population after rejecting and variation calculates;
Current genetic algebra is added 1;
Judge whether to reach default genetic algebra, if so, stopping heredity, finds out fitness value in current second population
Maximum individual determines that the value of the corresponding regular coefficient of this individual and kernel functional parameter is optimal regular coefficient and core
The value of function parameter is fitted if it is not, then returning to the execution fitness value of being rejected from second population less than preset first
The individual for answering angle value, until judging that obtaining current genetic algebra is equal to default genetic algebra.
Preferably, the kernel function in the least square method core extreme learning machine is Quadratic Rational kernel function.
Preferably, it is also wrapped before the picture in sample graph valut being input in recurrent least square method core extreme learning machine
It includes:
Image data in the sample graph valut is normalized.
A kind of sorter based on genetic optimization and core extreme learning machine, including:
First input unit, is used for:Picture in sample graph valut is input to the study of the recurrent least square method core limit
In machine;
Acquiring unit is used for:The recurrent least square method core extreme learning machine is obtained in multigroup different normalization system
The nicety of grading classified to the picture in the sample graph valut under number and kernel functional parameter, wherein one group of normalization system
A number nicety of grading corresponding with kernel functional parameter, the recurrent least square method core extreme learning machine is to the sample graph valut
In the classification results of picture include that the user of preset quantity likes grade;
Optimize unit, is used for:The regular coefficient is optimized with kernel functional parameter based on genetic algorithm, is obtained most
Excellent regular coefficient and kernel functional parameter, wherein any group of normalization coefficient nicety of grading corresponding with kernel functional parameter is
Fitness value in genetic algorithm;
Second input unit, is used for:The value of the optimal regular coefficient and kernel functional parameter is input to described pass
Return in least square method core extreme learning machine, the picture in picture library to be sorted is classified;
Recommendation unit is used for:Grade is liked to recommend from high to low to user according to corresponding user in picture.
Preferably, the optimization unit includes:
Generation unit is used for:Using multigroup different regular coefficient and kernel functional parameter as being planted in genetic algorithm
The individual of group, all individuals form the first group;Second with default individual amount is randomly generated from first population
Population;
Selecting unit is used for:That fitness value is less than preset first fitness value is rejected from second population
Body;
Intersect and become anticoincidence unit, is used for:Calculated crosswise is carried out to first population after rejecting and variation calculates;
Counting unit is used for:Current genetic algebra is added 1;
Judging unit is used for:Judge whether to reach default genetic algebra, if so, stopping heredity, finds out current second
The maximum individual of fitness value in population determines that the value of the corresponding regular coefficient of this individual and kernel functional parameter is optimal
The value of regular coefficient and kernel functional parameter, if it is not, it is small then to return to the execution rejecting fitness value from second population
In the individual of preset first fitness value, until judging that obtaining current genetic algebra is equal to default genetic algebra.
Preferably, the kernel function in the least square method core extreme learning machine is Quadratic Rational kernel function.
Preferably, further include:
Normalized unit, is used for:The picture in sample graph valut is input to recurrence in first input unit
Before in least square method core extreme learning machine, the image data in the sample graph valut is normalized.
A kind of sorting device based on genetic optimization and core extreme learning machine, including:
Memory is used for:Store computer program;
Processor is used for:It is realized when executing the computer program as described in any one of the above embodiments a kind of excellent based on heredity
The step of changing the sorting technique with core extreme learning machine.
A kind of computer readable storage medium is stored with computer program on the computer readable storage medium, described
It is realized when computer program is executed by processor as right is described in any one of the above embodiments a kind of based on genetic optimization and the core limit
The step of sorting technique of habit machine.
A kind of sorting technique based on genetic optimization and core extreme learning machine provided by the invention, including:By samples pictures
Picture in library is input in recurrent least square method core extreme learning machine;Obtain the recurrent least square method core limit study
The classification that machine classifies to the picture in the sample graph valut under multigroup different regular coefficient and kernel functional parameter
Precision, wherein one group of normalization coefficient nicety of grading corresponding with kernel functional parameter, the recurrent least square method core limit
Learning machine likes grade to the user that the classification results of the picture in the sample graph valut include preset quantity;It is calculated based on heredity
Method optimizes the regular coefficient with kernel functional parameter, obtains optimal regular coefficient and kernel functional parameter, wherein
Any group of normalization coefficient nicety of grading corresponding with kernel functional parameter is the fitness value in genetic algorithm;It will be described optimal
The value of regular coefficient and kernel functional parameter is input in the recurrent least square method core extreme learning machine, by figure to be sorted
Picture in valut is classified;Grade is liked to recommend from high to low to user according to corresponding user in picture.In this method
Sorting parameter in recurrent least square method core extreme learning machine be based on genetic algorithm optimization after obtain optimal classification ginseng
Number, nicety of grading are improved, and then are improved picture and recommended accuracy, are further promoted user experience and are increased user's viscosity.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow of the sorting technique based on genetic optimization and core extreme learning machine provided in an embodiment of the present invention
Figure;
Fig. 2 is a kind of structure of the sorter based on genetic optimization and core extreme learning machine provided in an embodiment of the present invention
Schematic diagram.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Referring to Fig. 1, it illustrates provided in an embodiment of the present invention a kind of based on genetic optimization and core extreme learning machine
Sorting technique may include:
Step S11:Picture in sample graph valut is input in recurrent least square method core extreme learning machine.
Picture in sample graph valut is for training recurrent least square method core extreme learning machine.
Step S12:Recurrent least square method core extreme learning machine is obtained in multigroup different regular coefficient and kernel function
The nicety of grading classified to the picture in sample graph valut under parameter, wherein one group of normalization coefficient and kernel functional parameter
A corresponding nicety of grading, recurrent least square method core extreme learning machine include to the classification results of the picture in sample graph valut
The user of preset quantity likes grade.
Determine the variation range of regular coefficient and kernel functional parameter first, then according to the variation range determined with
Machine generates multigroup regular coefficient and kernel functional parameter with concrete numerical value, by different groups of other regular coefficients and kernel function
Parameter is updated to correspondence in recurrent least square method core extreme learning machine and obtains the different recurrent least square method core limit respectively
Then the model of learning machine respectively classifies to the picture in sample graph valut with different models.It is input to recurrence minimum
The Standard User of picture in square law core extreme learning machine likes that grade is to predefine well, therefore according to above-mentioned different
Classification results and predetermined mark of the model of recurrent least square method core extreme learning machine to the picture in sample graph valut
Grade is liked to compare the nicety of grading that can obtain each model in mutatis mutandis family.User likes grade that can be drawn according to actual conditions
Point, such as can be divided into:Be delithted with, like, generally, do not like etc..
Step S13:Regular coefficient is optimized with kernel functional parameter based on genetic algorithm, obtains optimal normalization
Coefficient and kernel functional parameter, wherein any group of normalization coefficient nicety of grading corresponding with kernel functional parameter is in genetic algorithm
Fitness value.
Genetic algorithm is based on natural selection and theory of heredity, by survival of the fittest rule during biological evolution and kind
The efficient global optimization approach that the random information exchanging mechanism of group's intrinsic stain body is combined.It can be true by genetic algorithm
Make optimal regular coefficient and kernel functional parameter.
Step S14:The value of optimal regular coefficient and kernel functional parameter is input to the recurrent least square method core limit
In learning machine, the picture in picture library to be sorted is classified.
After the value for determining optimal regular coefficient and kernel functional parameter, so that it may optimal value is input to recurrence minimum
Optimal recurrent least square method core extreme learning machine model is obtained in square law core extreme learning machine, which is exactly to pass through sample
Then the trained model of picture in this picture library treats the picture in classification chart valut using the model and classifies.
Step S15:Grade is liked to recommend from high to low to user according to corresponding user in picture.
Picture likes higher grade to illustrate that user more likes the picture, therefore first will should like grade is high to push away when recommendation
It recommends to user, the rear recommendation for liking grade low to user.
Sorting parameter in this method in recurrent least square method core extreme learning machine is based on after genetic algorithm optimization
Obtained optimal classification parameter, nicety of grading are improved, and then are improved picture and recommended accuracy, and user experience is further promoted
Increase user's viscosity.
Preferably, optimized with kernel functional parameter to regular coefficient based on genetic algorithm, obtain it is optimal just
Ruleization coefficient includes with kernel functional parameter:
Using multigroup different regular coefficient and kernel functional parameter as the individual of population in genetic algorithm, all group of individuals
At the first population;
The second population with default individual amount is randomly generated from the first population;
The individual that fitness value is less than preset first fitness value is rejected from the second population;
Calculated crosswise is carried out to the first population after rejecting and variation calculates;
Current genetic algebra is added 1;
Judge whether to reach default genetic algebra, if so, stopping heredity, finds out fitness value in current second population
Maximum individual determines that the value of the corresponding regular coefficient of this individual and kernel functional parameter is optimal regular coefficient and core
The value of function parameter rejects fitness value less than preset first fitness value if it is not, then returning and executing from the second population
Individual, until judging that obtaining current genetic algebra is equal to default genetic algebra.
First population is made of total sample data, randomly generates out the second population from the first population, in the second population
Individual amount should be configured according to actual conditions, such as:The size of data volume to be sorted is considered, if the number that fruit is to be sorted
It is very big according to measuring, then can the quantity of the individual of the second population be arranged larger, it is on the contrary then can be arranged smaller.In reality
During border is realized, usually first the individual in the first population is encoded, coding mode can be carried out according to actual conditions
Setting, generally use binary coding, code length should be configured according to real data size.Calculated crosswise includes that single-point is handed over
Fork calculating, uniform crossover, intersection of shuffling, is reduced and acts on behalf of calculated crosswise etc. multiple-spot detection calculating, can be carried out according to actual conditions
Selection.As soon as often completing the rejecting in time above process, calculated crosswise and variation to calculate, primary heredity, hereditary result are completed
Heredity is obtained not less than the individual of the first fitness value and generate new more preferably individual for fitness value.Multi-Objective Genetic algebraically
It presets, can stop heredity if genetic algebra reaches preset value, otherwise continue heredity.
Preferably, the kernel function in least square method core extreme learning machine is Quadratic Rational kernel function.
Acting on for Quadratic Rational kernel function is very wide, can improve the classification speed of least square method core extreme learning machine,
Reduce time loss.
Preferably, before the picture in sample graph valut is input in recurrent least square method core extreme learning machine also
May include:Image data in sample graph valut is normalized.
After normalized, between the ranges of data all falls within 0 to 1, the recurrent least square method core limit can be improved
The processing speed of habit machine.
Referring to Fig. 2, it illustrates provided in an embodiment of the present invention a kind of based on genetic optimization and core extreme learning machine
Sorter may include:
First input unit 11, is used for:Picture in sample graph valut is input to the recurrent least square method core limit
In habit machine;
Acquiring unit 12, is used for:Recurrent least square method core extreme learning machine is obtained in multigroup different regular coefficient
With the nicety of grading classified to the picture in sample graph valut under kernel functional parameter, wherein one group of normalization coefficient and core
Function parameter corresponds to a nicety of grading, classification of the recurrent least square method core extreme learning machine to the picture in sample graph valut
As a result the user for including preset quantity likes grade;
Optimize unit 13, is used for:Regular coefficient is optimized with kernel functional parameter based on genetic algorithm, is obtained optimal
Regular coefficient and kernel functional parameter, wherein any group of normalization coefficient nicety of grading corresponding with kernel functional parameter be lose
Fitness value in propagation algorithm;
Second input unit 14, is used for:The value of optimal regular coefficient and kernel functional parameter is input to recurrence minimum
In square law core extreme learning machine, the picture in picture library to be sorted is classified;
Recommendation unit 15, is used for:Grade is liked to recommend from high to low to user according to corresponding user in picture.
Preferably, optimization unit 13 may include:
Generation unit is used for:Using multigroup different regular coefficient and kernel functional parameter as population in genetic algorithm
Individual, all individuals form the first group;The second population with default individual amount is randomly generated from the first population;
Selecting unit is used for:The individual that fitness value is less than preset first fitness value is rejected from the second population;
Intersect and become anticoincidence unit, is used for:Calculated crosswise is carried out to the first population after rejecting and variation calculates;
Counting unit is used for:Current genetic algebra is added 1;
Judging unit is used for:Judge whether to reach default genetic algebra, if so, stopping heredity, finds out current second
The maximum individual of fitness value in population determines that the value of the corresponding regular coefficient of this individual and kernel functional parameter is optimal
The value of regular coefficient and kernel functional parameter, if it is not, then return to execution rejects fitness value less than preset from the second population
The individual of first fitness value, until judging that obtaining current genetic algebra is equal to default genetic algebra.
Preferably, the kernel function in least square method core extreme learning machine is Quadratic Rational kernel function.
Preferably, a kind of sorter based on genetic optimization and core extreme learning machine provided in an embodiment of the present invention
Can also include:
Normalized unit, is used for:The picture in sample graph valut is input to recurrence minimum in the first input unit
Before in square law core extreme learning machine, the image data in sample graph valut is normalized.
Relevant portion in a kind of sorter based on genetic optimization and core extreme learning machine provided in an embodiment of the present invention
Explanation refer in a kind of sorting technique based on genetic optimization and core extreme learning machine provided in an embodiment of the present invention it is corresponding
Partial detailed description, details are not described herein.
The embodiment of the present invention additionally provides a kind of sorting device based on genetic optimization and core extreme learning machine, including:
Memory is used for:Store computer program;
Processor is used for:Realize that above-mentioned one one kind such as is based on genetic optimization and the core limit when executing computer program
The step of sorting technique of learning machine.
The present invention also provides a kind of computer readable storage medium, computer is stored on computer readable storage medium
Program realizes a kind of based on genetic optimization and core extreme learning machine of any of the above-described when computer program is executed by processor
The step of sorting technique.
In addition, in above-mentioned technical proposal provided in an embodiment of the present invention with correspond to technical solution realization principle in the prior art
Consistent part is simultaneously unspecified, in order to avoid excessively repeat.
The foregoing description of the disclosed embodiments enables those skilled in the art to realize or use the present invention.To this
A variety of modifications of a little embodiments will be apparent for a person skilled in the art, and the general principles defined herein can
Without departing from the spirit or scope of the present invention, to realize in other embodiments.Therefore, the present invention will not be limited
It is formed on the embodiments shown herein, and is to fit to consistent with the principles and novel features disclosed in this article widest
Range.
It should also be noted that, in the present specification, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only that
A little elements, but also include other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Claims (10)
1. a kind of sorting technique based on genetic optimization and core extreme learning machine, which is characterized in that including:
Picture in sample graph valut is input in recurrent least square method core extreme learning machine;
Obtain the recurrent least square method core extreme learning machine multigroup different regular coefficient with it is right under kernel functional parameter
The nicety of grading that picture in the sample graph valut is classified, wherein one group of normalization coefficient is corresponding with kernel functional parameter
One nicety of grading, classification results of the recurrent least square method core extreme learning machine to the picture in the sample graph valut
User including preset quantity likes grade;
The regular coefficient is optimized with kernel functional parameter based on genetic algorithm, obtains optimal regular coefficient and core
Function parameter, wherein any group of normalization coefficient nicety of grading corresponding with kernel functional parameter is the fitness in genetic algorithm
Value;
The value of the optimal regular coefficient and kernel functional parameter is input to the recurrent least square method core limit study
In machine, the picture in picture library to be sorted is classified;
Grade is liked to recommend from high to low to user according to corresponding user in picture.
2. sorting technique according to claim 1, which is characterized in that the genetic algorithm that is based on is to the regular coefficient
It is optimized with kernel functional parameter, obtain optimal regular coefficient includes with kernel functional parameter:
Using multigroup different regular coefficient and kernel functional parameter as the individual of population in genetic algorithm, all group of individuals
At the first population;
The second population with default individual amount is randomly generated from first population;
The individual that fitness value is less than preset first fitness value is rejected from second population;
Calculated crosswise is carried out to first population after rejecting and variation calculates;
Current genetic algebra is added 1;
Judge whether to reach default genetic algebra, if so, stopping heredity, it is maximum to find out fitness value in current second population
Individual, determine that the value of the corresponding regular coefficient of this individual and kernel functional parameter is optimal regular coefficient and kernel function
The value of parameter, if it is not, then returning to the execution fitness value of being rejected from second population is less than preset first fitness
The individual of value, until judging that obtaining current genetic algebra is equal to default genetic algebra.
3. method according to claim 1 or 2, which is characterized in that the core in the least square method core extreme learning machine
Function is Quadratic Rational kernel function.
4. according to the method described in claim 3, it is characterized in that, the picture in sample graph valut is input to recurrence minimum two
Further include before in multiplication core extreme learning machine:
Image data in the sample graph valut is normalized.
5. a kind of sorter based on genetic optimization and core extreme learning machine, which is characterized in that including:
First input unit, is used for:Picture in sample graph valut is input in recurrent least square method core extreme learning machine;
Acquiring unit is used for:Obtain the recurrent least square method core extreme learning machine multigroup different regular coefficient with
The nicety of grading classified to the picture in the sample graph valut under kernel functional parameter, wherein one group of normalization coefficient with
Kernel functional parameter corresponds to a nicety of grading, and the recurrent least square method core extreme learning machine is in the sample graph valut
The classification results of picture include that the user of preset quantity likes grade;
Optimize unit, is used for:The regular coefficient is optimized with kernel functional parameter based on genetic algorithm, is obtained optimal
Regular coefficient and kernel functional parameter, wherein any group of normalization coefficient nicety of grading corresponding with kernel functional parameter is heredity
Fitness value in algorithm;
Second input unit, is used for:The value of the optimal regular coefficient and kernel functional parameter is input to the recurrence most
In small square law core extreme learning machine, the picture in picture library to be sorted is classified;
Recommendation unit is used for:Grade is liked to recommend from high to low to user according to corresponding user in picture.
6. sorter according to claim 5, which is characterized in that the optimization unit includes:
Generation unit is used for:Using multigroup different regular coefficient and kernel functional parameter as population in genetic algorithm
Individual, all individuals form the first group;The second population with default individual amount is randomly generated from first population;
Selecting unit is used for:The individual that fitness value is less than preset first fitness value is rejected from second population;
Intersect and become anticoincidence unit, is used for:Calculated crosswise is carried out to first population after rejecting and variation calculates;
Counting unit is used for:Current genetic algebra is added 1;
Judging unit is used for:Judge whether to reach default genetic algebra, if so, stopping heredity, finds out current second population
The middle maximum individual of fitness value determines that the value of the corresponding regular coefficient of this individual and kernel functional parameter is optimal regular
Change the value of coefficient and kernel functional parameter, if it is not, then returning to the execution fitness value of being rejected from second population less than pre-
If the first fitness value individual, until judging that obtaining current genetic algebra is equal to default genetic algebra.
7. device according to claim 5 or 6, which is characterized in that the core in the least square method core extreme learning machine
Function is Quadratic Rational kernel function.
8. device according to claim 7, which is characterized in that further include:
Normalized unit, is used for:The picture in sample graph valut is input to recurrence minimum in first input unit
Before in square law core extreme learning machine, the image data in the sample graph valut is normalized.
9. a kind of sorting device based on genetic optimization and core extreme learning machine, which is characterized in that including:
Memory is used for:Store computer program;
Processor is used for:Realize that Claims 1-4 any one of them one kind such as is based on losing when executing the computer program
The step of passing optimization and the sorting technique of core extreme learning machine.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program is realized as excellent based on heredity such as Claims 1-4 any one of them one kind when the computer program is executed by processor
The step of changing the sorting technique with core extreme learning machine.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810295090.XA CN108664992B (en) | 2018-03-30 | 2018-03-30 | Classification method and device based on genetic optimization and kernel extreme learning machine |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810295090.XA CN108664992B (en) | 2018-03-30 | 2018-03-30 | Classification method and device based on genetic optimization and kernel extreme learning machine |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108664992A true CN108664992A (en) | 2018-10-16 |
CN108664992B CN108664992B (en) | 2022-02-15 |
Family
ID=63783069
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810295090.XA Active CN108664992B (en) | 2018-03-30 | 2018-03-30 | Classification method and device based on genetic optimization and kernel extreme learning machine |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108664992B (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103400190A (en) * | 2013-08-13 | 2013-11-20 | 浙江大学 | Integrated framework method for optimizing extremity learning machine by using genetic algorithm |
CN104537382A (en) * | 2015-01-12 | 2015-04-22 | 杭州电子科技大学 | Electromyographic signal gait recognition method for optimizing support vector machine based on genetic algorithm |
CN104657718A (en) * | 2015-02-13 | 2015-05-27 | 武汉工程大学 | Face recognition method based on face image feature extreme learning machine |
CN104700078A (en) * | 2015-02-13 | 2015-06-10 | 武汉工程大学 | Scale-invariant feature extreme learning machine-based robot scene recognition method |
CN105718963A (en) * | 2016-03-09 | 2016-06-29 | 东南大学 | SAR image classification method based on variable-length incremental type extreme learning machine |
US9785886B1 (en) * | 2017-04-17 | 2017-10-10 | SparkCognition, Inc. | Cooperative execution of a genetic algorithm with an efficient training algorithm for data-driven model creation |
CN107679543A (en) * | 2017-02-22 | 2018-02-09 | 天津大学 | Sparse autocoder and extreme learning machine stereo image quality evaluation method |
CN107798383A (en) * | 2017-10-27 | 2018-03-13 | 天津大学 | Improved core extreme learning machine localization method |
-
2018
- 2018-03-30 CN CN201810295090.XA patent/CN108664992B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103400190A (en) * | 2013-08-13 | 2013-11-20 | 浙江大学 | Integrated framework method for optimizing extremity learning machine by using genetic algorithm |
CN104537382A (en) * | 2015-01-12 | 2015-04-22 | 杭州电子科技大学 | Electromyographic signal gait recognition method for optimizing support vector machine based on genetic algorithm |
CN104657718A (en) * | 2015-02-13 | 2015-05-27 | 武汉工程大学 | Face recognition method based on face image feature extreme learning machine |
CN104700078A (en) * | 2015-02-13 | 2015-06-10 | 武汉工程大学 | Scale-invariant feature extreme learning machine-based robot scene recognition method |
CN105718963A (en) * | 2016-03-09 | 2016-06-29 | 东南大学 | SAR image classification method based on variable-length incremental type extreme learning machine |
CN107679543A (en) * | 2017-02-22 | 2018-02-09 | 天津大学 | Sparse autocoder and extreme learning machine stereo image quality evaluation method |
US9785886B1 (en) * | 2017-04-17 | 2017-10-10 | SparkCognition, Inc. | Cooperative execution of a genetic algorithm with an efficient training algorithm for data-driven model creation |
CN107798383A (en) * | 2017-10-27 | 2018-03-13 | 天津大学 | Improved core extreme learning machine localization method |
Also Published As
Publication number | Publication date |
---|---|
CN108664992B (en) | 2022-02-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Azzeh et al. | An empirical evaluation of ensemble adjustment methods for analogy-based effort estimation | |
CN105138653B (en) | It is a kind of that method and its recommendation apparatus are recommended based on typical degree and the topic of difficulty | |
CN111242310B (en) | Feature validity evaluation method and device, electronic equipment and storage medium | |
CN107301225A (en) | Short text classification method and device | |
CN112541532B (en) | Target detection method based on dense connection structure | |
CN110084271A (en) | A kind of other recognition methods of picture category and device | |
CN112258093A (en) | Risk level data processing method and device, storage medium and electronic equipment | |
CN105354595A (en) | Robust visual image classification method and system | |
CN104050197B (en) | A kind of information retrieval system evaluating method and device | |
CN110737859A (en) | UP main matching method and device | |
CN115358809A (en) | Multi-intention recommendation method and device based on graph comparison learning | |
CN114168761A (en) | Multimedia data pushing method and device, electronic equipment and storage medium | |
CN105989066A (en) | Information processing method and device | |
CN107133811A (en) | The recognition methods of targeted customer a kind of and device | |
CN106776757A (en) | User completes the indicating means and device of Net silver operation | |
CN108304568A (en) | A kind of real estate Expectations big data processing method and system | |
CN112132589A (en) | Method for constructing fraud recognition model based on multiple times of fusion | |
CN110059749B (en) | Method and device for screening important features and electronic equipment | |
CN108388652B (en) | Method and device for sending song list identification | |
CN108664992A (en) | A kind of sorting technique and device based on genetic optimization and core extreme learning machine | |
CN114676634A (en) | Water level prediction method and system based on NSGA-III optimized LSTM neural network | |
CN108229572A (en) | A kind of parameter optimization method and computing device | |
CN112231930B (en) | Distributed hydrological model parameter optimization method | |
CN113537461A (en) | Network key node discovery method and system based on SIR value learning | |
CN116244426A (en) | Geographic function area identification method, device, equipment and storage medium |
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 |