CN114742177A - Meteorological data classification method based on AGA-XGboost and GWO-SVM - Google Patents

Meteorological data classification method based on AGA-XGboost and GWO-SVM Download PDF

Info

Publication number
CN114742177A
CN114742177A CN202210639860.4A CN202210639860A CN114742177A CN 114742177 A CN114742177 A CN 114742177A CN 202210639860 A CN202210639860 A CN 202210639860A CN 114742177 A CN114742177 A CN 114742177A
Authority
CN
China
Prior art keywords
meteorological data
xgboost
gwo
aga
svm
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
Application number
CN202210639860.4A
Other languages
Chinese (zh)
Other versions
CN114742177B (en
Inventor
秦华旺
尹传豪
戴跃伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Information Science and Technology
Original Assignee
Nanjing University of Information Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University of Information Science and Technology filed Critical Nanjing University of Information Science and Technology
Priority to CN202210639860.4A priority Critical patent/CN114742177B/en
Publication of CN114742177A publication Critical patent/CN114742177A/en
Application granted granted Critical
Publication of CN114742177B publication Critical patent/CN114742177B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2431—Multiple classes
    • 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
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • 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
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Business, Economics & Management (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Tourism & Hospitality (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Biophysics (AREA)
  • Development Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Educational Administration (AREA)
  • General Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • Economics (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Complex Calculations (AREA)

Abstract

The invention discloses a meteorological data classification method based on AGA-XGboost and GWO-SVM, which comprises the following steps: preprocessing meteorological data; constructing a stacking noise reduction self-encoder, and extracting the characteristics of the preprocessed training set data; using the processed training set sample as a learning sample of two classifier models of AGA-XGboost and GWO-SVM; and selecting a classifier model suitable for the sample to be classified according to the proportion of each category on the selected meteorological data set. The meteorological data classification method can extract deep features from massive high-dimensional meteorological data, effectively solve the problem of data imbalance in meteorological samples and improve the overall classification accuracy of the model.

Description

Meteorological data classification method based on AGA-XGboost and GWO-SVM
Technical Field
The invention belongs to the technical field of meteorological data classification, and particularly relates to a meteorological data classification method based on AGA-XGboost and GWO-SVM.
Background
With the continuous development of meteorological observation technology, the meteorological data generated every day is multiplied, and great challenges are generated for inquiring and processing the meteorological data, so that the method for searching the meteorological data with high classification accuracy and stability has very important significance.
The meteorological data comprise a great amount of meteorological element information, the frequency of certain extreme weather is very low, the sample data size difference of different types is too large, the traditional classification method aims at maximizing the accuracy of integral classification, the classification result is biased to a plurality of types, the classification of the meteorological data in the prior art is mostly binary classification aiming at a single type, and the research of multi-classification of a plurality of meteorological types can be rarely carried out at the same time.
Therefore, how to reasonably establish a multi-classification model of meteorological data is a technical difficulty in improving the classification accuracy of small samples on the premise of ensuring the classification accuracy of large sample data in the meteorological data.
Disclosure of Invention
The technical problem to be solved is as follows: aiming at the technical problems, the invention provides a meteorological data classification method based on AGA-XGboost and GWO-SVM, and solves the problems of poor classification effect, poor model self-adaption capability and unbalanced data of high-dimensional meteorological data of the sea under the existing classification method.
The technical scheme is as follows: the meteorological data classification method based on the AGA-XGboost and the GWO-SVM comprises the following steps:
s1, meteorological data preprocessing:
s1-1, replacing abnormal values in the meteorological data and filling default values in the meteorological data;
s1-2, performing one-hot encoding on meteorological data;
s1-3, carrying out min-max normalization processing on different types of meteorological data to obtain mapping data between [0-1 ];
s1-4, judging whether a Borderline-SMOTE method is adopted for oversampling according to a meteorological data processing result;
s2, constructing a stacking noise reduction self-encoder, and extracting the characteristics of the preprocessed training set data;
s3, constructing two classifier models of AGA-XGboost and GWO-SVM, and selecting different classifier models according to the proportion of different class attributes of samples in a training set in the selected meteorological data set;
and S4, passing the preprocessed sample to be classified through a constructed classifier model to obtain a meteorological data classification result.
Preferably, the default values are filled in the step S1-1 by linear interpolation or mean smoothing.
Preferably, the categories of the classification of the meteorological data in the step S1-2 include sunny days, rainy days, cloudy days, snowy days, foggy days and hail days.
Preferably, the processing formula of min-max normalization in step S1-3 is:
Figure 385967DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 21217DEST_PATH_IMAGE002
is a value normalized by single-type meteorological data,
Figure 66533DEST_PATH_IMAGE003
is the minimum value of the single-type meteorological data,
Figure 930584DEST_PATH_IMAGE004
is the maximum value of the single-type meteorological data.
Preferably, the judgment criteria in step S1-4 are as follows: when the ratio of the number of the minority class samples to the number of the majority class samples is smaller than the set unbalance tolerance, performing Borderline-SMOTE for oversampling; otherwise, the operation is not executed.
Preferably, the step S2 includes the following steps:
s2-1, using the training set data as the input of the first DA unit of the stacking noise reduction self-encoder;
s2-2, updating the weight and the bias by using a gradient descent method by using the minimum variance as a cost function;
s2-3, removing the output layer and the corresponding weight and bias, only keeping the weight and bias of the input layer and the hidden layer, then using the hidden layer of the first DA unit as the input of the second DA unit, and so on, training layer by layer.
Preferably, the AGA-XGboost classifier model is constructed by the following method: and establishing an initialization population, calculating the fitness value of each individual, judging whether a stopping condition is met, if the stopping condition is not met or the target progress is not met, sequentially performing selection, crossing and mutation operations until the stopping condition is met, outputting the optimal parameter group of the XGboost, and obtaining the optimized AGA-XGboost classifier model.
Preferably, the selecting operation is: using the roulette method, fitness values for each individual are calculated to form a roulette scale for random selection.
Preferably, the interleaving operation is: a threshold k is established, crossing at the kth position of the gene.
Preferably, the mutation operation is: selecting the gene mutation position according to the mutation probability, performing 0-1 conversion to prevent local optimization, updating the values of the cross probability a and the mutation probability b according to the following formula, and accelerating the iteration speed:
Figure 213798DEST_PATH_IMAGE005
wherein
Figure 505102DEST_PATH_IMAGE006
And
Figure 991709DEST_PATH_IMAGE007
respectively the maximum and minimum of the cross-over probability,
Figure 393872DEST_PATH_IMAGE008
respectively the maximum and minimum of the probability of variation,
Figure 531592DEST_PATH_IMAGE009
is the individual fitness value in the population of the i generation,
Figure 790535DEST_PATH_IMAGE010
is the average value of fitness in the population of the i generation.
Preferably, the GWO-SVM classifier model is constructed as follows: setting a kernel function and a penalty factor of the SVM, initializing parameters of the gray wolf algorithm, judging whether a termination condition is reached, if not, calculating population fitness, keeping individuals of the first three fitness, sequentially updating the positions of the rest gray wolfs, updating the parameters of the gray wolf algorithm until the termination condition is reached, and obtaining an optimized GWO-SVM classifier model.
Has the advantages that: the method extracts deep features of the meteorological data by using the stacked noise reduction self-encoder, reduces the influence of an unbalanced meteorological data set on the classification effect by using an oversampling algorithm, and makes up the defects of the traditional classifier by using the training classifier which can be coupled in a self-adaptive manner according to different samples.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a flow chart of Borderline-SMOTE oversampling;
FIG. 3 is a flow chart of the construction of a classifier model.
Detailed Description
The invention is further described below with reference to the accompanying drawings and specific embodiments.
Example 1
As shown in FIGS. 1 to 3, a meteorological data classification method based on AGA-XGboost and GWO-SVM comprises the following steps:
s1, meteorological data preprocessing:
s1-1, replacing abnormal values in the meteorological data, filling default values in the meteorological data:
processing abnormal values and default values in the meteorological data, for example, when missing values appear in the atmospheric pressure data, selecting the atmospheric pressure in the near time for replacement, and selecting a mean value smoothing method for replacement according to the relative humidity;
s1-2, carrying out one-hot coding on meteorological data:
and carrying out one-bit independent thermal coding on weather types in the meteorological data set, such as sunny days, rainy days, cloudy days, foggy days, snowing days, thunderstorms, hail days and the like. In this embodiment, the meteorological data are divided into 6 categories, labeled, and mapped to 000001 in sunny days, 000010 in rainy days, 000100 in cloudy days, 001000 in snowy days, 010000 in hail days, and 100000 in foggy days;
s1-3, carrying out min-max normalization processing on different types of meteorological data to obtain mapping data between [0-1 ]:
due to the difference of dimension units and magnitude among a plurality of meteorological data, the original data needs to be linearly transformed, the transformed data are all mapped to [0-1], and the transformation formula is as follows:
Figure 13706DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 468827DEST_PATH_IMAGE002
is a value normalized by single-type meteorological data,
Figure 195475DEST_PATH_IMAGE003
is the minimum value of the single-type meteorological data,
Figure 890898DEST_PATH_IMAGE004
the maximum value of the single-type meteorological data;
s1-4, judging whether to adopt a Borderline-SMOTE method to perform oversampling according to the meteorological data processing result, and specifically comprising the following steps:
(1) let the number of samples in the training set samples be
Figure 601365DEST_PATH_IMAGE011
Each sample has a class label, in
Figure 345330DEST_PATH_IMAGE012
=4, 5 represents the minority samples and the majority samples, the minority samples represent some unusual extreme weather in the meteorological data set, and the number of the minority samples is set as
Figure 211305DEST_PATH_IMAGE013
The number of majority class samples is msAnd is and
Figure 15313DEST_PATH_IMAGE014
if g is<
Figure 947497DEST_PATH_IMAGE015
Then Borderline-SMOTE is executed, wherein
Figure 495153DEST_PATH_IMAGE015
Maximum imbalance tolerance;
(2) and (4) taking the sample set of the minority samples as P, calculating the Euclidean distance between each sample in the P and the sample of the whole training set, searching the neighbor of the sample, wherein the neighbor number is k, and the samples belonging to the majority are marked as
Figure 258709DEST_PATH_IMAGE016
If, if
Figure 748465DEST_PATH_IMAGE016
If k, judging the sample as noise; if k/2<
Figure 902366DEST_PATH_IMAGE016
<k, judging the sample as a boundary sample; if it is
Figure 253713DEST_PATH_IMAGE017
Then it is a safety sample.
(3) Synthesizing the boundary sample, wherein the synthesis formula is as follows:
Figure 75039DEST_PATH_IMAGE018
wherein
Figure 283166DEST_PATH_IMAGE019
In order to create a new sample of the sample,
Figure 940675DEST_PATH_IMAGE020
for the few classes of samples that are at the boundary,
Figure 564554DEST_PATH_IMAGE021
the random is the neighbor of a few samples of the boundary, and the rand generates a random number between 0 and 1.
S2, constructing a stacking noise reduction self-encoder, and performing feature extraction on the preprocessed training set data:
s2-1, using the training set data as input to the first DA cell of the stacked noise reduction autocode SDAE, the formula is as follows:
Figure 505965DEST_PATH_IMAGE022
wherein
Figure 88256DEST_PATH_IMAGE023
Is the sigmoid function, y is the data after adding noise,
Figure 279066DEST_PATH_IMAGE024
for recovered data or features, b1Coding bias for DA1, b2A decoding bias of DA 1; w1Is a weight matrix of the DA1,
Figure 221483DEST_PATH_IMAGE025
is the transpose of the weight matrix.
S2-2, adopting the minimum variance as a cost function, and utilizing a gradient descent method to update the weight W and the bias b, wherein the formula is as follows:
Figure 751822DEST_PATH_IMAGE026
；
s2-3, removing the output layer and its corresponding weight and bias, and only keeping the weight W of the input layer and the hidden layer1And bias b1And then, taking the hidden layer of the first DA unit as the input of the second DA unit, and so on, training layer by layer.
S3, constructing two classifier models of AGA-XGboost and GWO-SVM, and selecting the classifier models according to the proportion of the sample size in the data set:
the construction method of the AGA-XGboost classifier model comprises the following steps: establishing an initialization population, calculating the fitness value of each individual, judging whether a stopping condition is met or not, if the maximum iteration number is not met or the target progress is not met, sequentially performing selection, crossing and mutation operations until the stopping condition is met, outputting the optimal parameter group of the XGboost, and obtaining the optimized AGA-XGboost classifier model.
Wherein the selecting operation is: using the roulette method, fitness values for each individual are calculated to form a roulette scale for random selection.
The cross operation is that: a threshold k is established, crossing at the kth position of the gene.
The mutation operation is: selecting the gene mutation position according to the mutation probability, performing 0-1 conversion to prevent local optimization, and updating the values of the cross probability a and the mutation probability b according to the following formula to accelerate the iteration speed:
Figure 505014DEST_PATH_IMAGE005
wherein
Figure 120803DEST_PATH_IMAGE006
And
Figure 414381DEST_PATH_IMAGE007
respectively the maximum and minimum of the cross-over probability,
Figure 812608DEST_PATH_IMAGE008
respectively the maximum value and the minimum value of the mutation probability,
Figure 471123DEST_PATH_IMAGE009
is the individual fitness value in the population of the i generation,
Figure 574208DEST_PATH_IMAGE010
is the average value of fitness in the population of the i generation.
The construction method of the GWO-SVM classifier model comprises the following steps: setting a kernel function and a penalty factor of the SVM, initializing parameters of the gray wolf algorithm, judging whether a termination condition is reached, if not, calculating population fitness, keeping individuals of the first three fitness, sequentially updating the positions of the rest gray wolfs, updating the parameters of the gray wolf algorithm until the termination condition is reached, and obtaining an optimized GWO-SVM classifier model.
And then, the meteorological data processed by SDAE are used as learning samples of two algorithms of AGA-XGboost and GWO-SVM, and an execution scheme is selected according to the percentage of different types of attributes of the samples in the training set in the selected meteorological data set.
And S4, passing the preprocessed sample to be classified through a constructed classifier model to obtain a meteorological data classification result.

Claims (10)

1. The meteorological data classification method based on the AGA-XGboost and the GWO-SVM is characterized by comprising the following steps of:
s1, meteorological data preprocessing:
s1-1, replacing abnormal values in the meteorological data and filling default values in the meteorological data;
s1-2, performing one-hot encoding on meteorological data;
s1-3, carrying out min-max normalization processing on different types of meteorological data to obtain mapping data between [0-1 ];
s1-4, judging whether to adopt a Borderline-SMOTE method to perform oversampling according to a meteorological data processing result;
s2, constructing a stacking noise reduction self-encoder, and extracting the characteristics of the preprocessed training set data;
s3, constructing two classifier models of AGA-XGboost and GWO-SVM, and selecting the classifier models according to the proportion of different class attributes of samples in a training set in a selected meteorological data set;
and S4, passing the preprocessed sample to be classified through a constructed classifier model to obtain a meteorological data classification result.
2. The AGA-XGboost and GWO-SVM based meteorological data classification method according to claim 1, wherein the step S1-1 adopts a linear interpolation method or a mean smoothing method to fill default values.
3. The AGA-XGboost and GWO-SVM based meteorological data classification method according to claim 1, wherein the categories of meteorological data classification in the step S1-2 include sunny days, rainy days, cloudy days, snowy days, foggy days and hail days.
4. The AGA-XGboost and GWO-SVM based meteorological data classification method according to claim 1, wherein the processing formula of min-max normalization in the step S1-3 is as follows:
Figure 134520DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 905030DEST_PATH_IMAGE002
is a normalized value of the single-category meteorological data,
Figure 734446DEST_PATH_IMAGE003
is the minimum value of the single-category meteorological data,
Figure 590406DEST_PATH_IMAGE004
is the maximum value of the single-category meteorological data.
5. The AGA-XGboost and GWO-SVM based meteorological data classification method according to claim 1, wherein the judgment criteria in the step S1-4 are as follows: when the ratio of the number of the minority class samples to the number of the majority class samples is smaller than the set unbalance tolerance, performing Borderline-SMOTE for oversampling; otherwise, the operation is not executed.
6. The AGA-XGboost and GWO-SVM based meteorological data classification method according to claim 1, wherein the step S2 comprises the following steps:
s2-1, using the training set data as the input of the first DA unit of the stacking noise reduction self-encoder;
s2-2, updating the weight and the bias by using a gradient descent method by using the minimum variance as a cost function;
s2-3, removing the output layer and the corresponding weight and bias, only keeping the weight and bias of the input layer and the hidden layer, then using the hidden layer of the first DA unit as the input of the second DA unit, and so on, training layer by layer.
7. The meteorological data classification method based on the AGA-XGboost and GWO-SVM according to claim 1, characterized in that the AGA-XGboost classifier model is constructed as follows: and establishing an initialization population, calculating the fitness value of each individual, judging whether a stopping condition is met or not, if the stopping condition is not met or the target precision is not met, sequentially performing selection, crossing and mutation operations until the stopping condition is met, outputting the optimal parameter group of the XGboost, and obtaining the optimized AGA-XGboost classifier model.
8. The AGA-XGboost and GWO-SVM based meteorological data classification method of claim 7, wherein the selecting is performed by: calculating the fitness value of each individual to form a roulette proportion for random selection by using a roulette method; the interleaving operation is: a threshold k is established, crossing at the kth position of the gene.
9. The AGA-XGboost and GWO-SVM based meteorological data classification method according to claim 7, wherein the mutation operation is: selecting the gene mutation position according to the mutation probability, performing 0-1 conversion to prevent local optimization, updating the values of the cross probability a and the mutation probability b according to the following formula, and accelerating the iteration speed:
Figure 678317DEST_PATH_IMAGE005
wherein
Figure 37754DEST_PATH_IMAGE006
And
Figure 100388DEST_PATH_IMAGE007
respectively the maximum and minimum of the cross-over probability,
Figure 178065DEST_PATH_IMAGE008
respectively the maximum and minimum of the probability of variation,
Figure 820399DEST_PATH_IMAGE009
is the individual fitness value in the population of the i generation,
Figure 782146DEST_PATH_IMAGE010
is the average value of fitness in the population of the i generation.
10. The AGA-XGboost and GWO-SVM based meteorological data classification method according to claim 1, wherein the GWO-SVM classifier model is constructed by the following method: setting a kernel function and a penalty factor of the SVM, initializing parameters of the gray wolf algorithm, judging whether a termination condition is reached, if not, calculating population fitness, keeping individuals of the first three fitness, sequentially updating the positions of the rest gray wolfs, updating the parameters of the gray wolf algorithm until the termination condition is reached, and obtaining an optimized GWO-SVM classifier model.
CN202210639860.4A 2022-06-08 2022-06-08 Meteorological data classification method based on AGA-XGboost and GWO-SVM Active CN114742177B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210639860.4A CN114742177B (en) 2022-06-08 2022-06-08 Meteorological data classification method based on AGA-XGboost and GWO-SVM

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210639860.4A CN114742177B (en) 2022-06-08 2022-06-08 Meteorological data classification method based on AGA-XGboost and GWO-SVM

Publications (2)

Publication Number Publication Date
CN114742177A true CN114742177A (en) 2022-07-12
CN114742177B CN114742177B (en) 2022-09-02

Family

ID=82287711

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210639860.4A Active CN114742177B (en) 2022-06-08 2022-06-08 Meteorological data classification method based on AGA-XGboost and GWO-SVM

Country Status (1)

Country Link
CN (1) CN114742177B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115794981A (en) * 2022-12-14 2023-03-14 广西电网有限责任公司 Method and system for counting meteorological data by using model
CN116842454A (en) * 2023-06-06 2023-10-03 南京财经大学 Financial asset classification method and system based on support vector machine algorithm

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104573013A (en) * 2015-01-09 2015-04-29 上海大学 Category weight combined integrated learning classifying method
CN108470187A (en) * 2018-02-26 2018-08-31 华南理工大学 A kind of class imbalance question classification method based on expansion training dataset
CN109711452A (en) * 2018-12-20 2019-05-03 四川新网银行股份有限公司 It is a kind of based on WGAN-GP model to the uneven classification method of user behavior
CN111612628A (en) * 2020-05-28 2020-09-01 深圳博普科技有限公司 Method and system for classifying unbalanced data sets
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

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104573013A (en) * 2015-01-09 2015-04-29 上海大学 Category weight combined integrated learning classifying method
CN108470187A (en) * 2018-02-26 2018-08-31 华南理工大学 A kind of class imbalance question classification method based on expansion training dataset
CN109711452A (en) * 2018-12-20 2019-05-03 四川新网银行股份有限公司 It is a kind of based on WGAN-GP model to the uneven classification method of user behavior
CN111612628A (en) * 2020-05-28 2020-09-01 深圳博普科技有限公司 Method and system for classifying unbalanced data sets
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

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
杨毅: "面向不平衡数据集的一种精化...rline-SMOTE方法", 《复旦学报》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115794981A (en) * 2022-12-14 2023-03-14 广西电网有限责任公司 Method and system for counting meteorological data by using model
CN115794981B (en) * 2022-12-14 2023-09-26 广西电网有限责任公司 Method and system for counting meteorological data by using model
CN116842454A (en) * 2023-06-06 2023-10-03 南京财经大学 Financial asset classification method and system based on support vector machine algorithm
CN116842454B (en) * 2023-06-06 2024-04-30 南京财经大学 Financial asset classification method and system based on support vector machine algorithm

Also Published As

Publication number Publication date
CN114742177B (en) 2022-09-02

Similar Documents

Publication Publication Date Title
CN114742177B (en) Meteorological data classification method based on AGA-XGboost and GWO-SVM
CN111797779A (en) Remote sensing image semantic segmentation method based on regional attention multi-scale feature fusion
CN108932314A (en) A kind of chrysanthemum image content retrieval method based on the study of depth Hash
CN109815886A (en) A kind of pedestrian and vehicle checking method and system based on improvement YOLOv3
CN105512289A (en) Image retrieval method based on deep learning and Hash
CN105320961A (en) Handwriting numeral recognition method based on convolutional neural network and support vector machine
CN109272139A (en) It is a kind of based on Nonlinear Set at the short-term wind speed forecasting method of deep learning
CN113936204B (en) High-resolution remote sensing image cloud and snow identification method and device fusing terrain data and deep neural network
CN111832812A (en) Wind power short-term prediction method based on deep learning
CN111681132B (en) Typical power consumption mode extraction method suitable for massive class unbalanced load data
CN102750286A (en) Novel decision tree classifier method for processing missing data
CN114792156A (en) Photovoltaic output power prediction method and system based on curve characteristic index clustering
CN114462718A (en) CNN-GRU wind power prediction method based on time sliding window
CN116128150B (en) Photovoltaic intelligent output prediction method based on two-stage optimization
CN114897129A (en) Photovoltaic power station short-term power prediction method based on similar daily clustering and Kmeans-GRA-LSTM
CN112001293A (en) Remote sensing image ground object classification method combining multi-scale information and coding and decoding network
CN113568068A (en) Strong convection weather prediction method based on MPI parallel three-dimensional neural network
CN112668633B (en) Adaptive graph migration learning method based on fine granularity field
CN116502074A (en) Model fusion-based photovoltaic power generation power prediction method and system
CN112434740A (en) Depth learning-based Qin tomb warriors fragment classification method
CN105760895B (en) Method for dividing photovoltaic output fluctuation types
CN111488974A (en) Deep learning neural network-based ocean wind energy downscaling method
CN115481788B (en) Phase change energy storage system load prediction method and system
CN114865620A (en) Wind power plant generating capacity prediction method based on machine learning algorithm
CN110738249B (en) Aurora image clustering method based on deep neural network

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