CN110516733A - A kind of Recognition of Weil Logging Lithology method based on the more twin support vector machines of classification of improvement - Google Patents
A kind of Recognition of Weil Logging Lithology method based on the more twin support vector machines of classification of improvement Download PDFInfo
- Publication number
- CN110516733A CN110516733A CN201910781790.4A CN201910781790A CN110516733A CN 110516733 A CN110516733 A CN 110516733A CN 201910781790 A CN201910781790 A CN 201910781790A CN 110516733 A CN110516733 A CN 110516733A
- Authority
- CN
- China
- Prior art keywords
- class
- sample
- data
- classification
- lithology
- 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 25
- 238000012706 support-vector machine Methods 0.000 title claims abstract description 19
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000005070 sampling Methods 0.000 claims abstract description 4
- 238000012360 testing method Methods 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000005457 optimization Methods 0.000 claims description 9
- 239000011435 rock Substances 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 5
- 230000002123 temporal effect Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000013480 data collection Methods 0.000 claims description 2
- 238000007635 classification algorithm Methods 0.000 abstract description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000009977 dual effect Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 208000035126 Facies Diseases 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000009394 selective breeding Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
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
- G06F18/214—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
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Development Economics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Animal Husbandry (AREA)
- Agronomy & Crop Science (AREA)
- Mining & Mineral Resources (AREA)
- Operations Research (AREA)
- Marine Sciences & Fisheries (AREA)
Abstract
The invention discloses a kind of based on the Lithology Identification Methods for improving more twin support vector machines of classification, Lithology Discrimination is the basis for recognizing stratum and solving reservoir parameter, it is influenced by geological environment complexity and heterogeneity, there is a large amount of information redundancy and data set distribution between class imbalance problem between log, criteria classification algorithm is unable to satisfy actual demand.For existing criteria classification algorithm poor fault tolerance, identification lithology is single and can not effectively overcome the problems, such as unbalanced between class, the present invention utilizes over-sampling data, more conducively trained the data set for improving more twin support vector machines of classification, it is proposed that a kind of new fuzzy membership function improves twin support vector machines, so that improved twin support vector machines is particularly suited for training log data, the accuracy rate of Lithology Discrimination is effectively increased.
Description
Technical field
It is in particular to a kind of based on the more twin support vector machines of classification of improvement the present invention relates to petroleum exploration domain
Recognition of Weil Logging Lithology method.
Background technique
Lithology Discrimination is one of important process of logging Reservoir Evaluation, and fine lithology classification is to determine the exhibition of target area space
The reliable basis of cloth.Rock core is the best source of lithological information, but since the availability of rock core is limited, therefore pass through well logging
Physics rock property is measured, is had very by there is the well logging information of core well (training well) to carry out prediction to the lithology for unminding well
Important meaning.But since depositional environment complexity causes log parameter distribution more fuzzy, various complicated rocks can not be clearly distinguished
Property, and a plurality of log response causes its correlation higher often there is a large amount of information redundancy.Moreover, real
Border geological condition is complicated, and a certain area may be mainly made of two or three kind of lithology, causes data set imbalance problem, to fine
Lithology classification bring huge challenge.Lithology Discrimination is a typical high dimensional nonlinear mode identification procedure.Criteria classification
Algorithm is designed to the distribution of balance class, usually can not effectively solve imbalance problem between class, lead to the class that data volume is small
Classifying quality is poor.Existing cluster, neural network and traditional support vector machine scheduling algorithm can not effectively overcome imbalance between class to ask
Topic.
Summary of the invention
The technical problem to be solved by the present invention is in order to overcome the shortcomings in the prior art, the present invention provides a kind of accurate
Rate is high based on the Recognition of Weil Logging Lithology method for improving more twin support vector machines of classification.
The technical solution adopted by the present invention is that: it is a kind of based on the Recognition of Weil Logging Lithology for improving more twin support vector machines of classification
Method, it is characterised in that the following steps are included:
Step 1: by data class sample over-sampling small in data set T, obtaining the new data set T' of all kinds of balances；
Step 2: new data set T' being pre-processed, and makes label, indicates that the label of k class lithology, k are with 0~k-1
New log data collection is divided into training set and test set by the positive integer more than or equal to 1；
Step 3: designing a kind of new fuzzy membership function, two classification that construction obscures twin support vector machines are super flat
The optimization method in face obscures twin support vector machines knot by " one-to-one " more classification policies and using new fuzzy membership function
It closes；
If matrixWithThe data sample of positive class and negative class is respectively indicated, wherein matrix A and B's is every
A line respectively indicates a data sample, m1Indicate the data sample number of positive class, m2Indicate that the sample number of negative class, n are data sample
Dimension, the center of positive classThe center of negative classDistance d of the positive class each point away from its central pointA=| | Ai-CA|
|, distance d of the negative class each point away from its central pointB=| | Bi-CB| |, positive class radius rA=max | Ai-CA|, negative class radius rB=max |
Bi-CB|,AiFor a certain sample, B in AiFor a certain sample in B；
For i, i=1,2 ... each in data set, m1+m2, the fuzzy membership S that is proposediIt is as follows:
δ=10 in formula-5；
Wherein, the optimization method of two hyperplane is as follows:
In formula, w(1)For the weight matrix of first kind sample, w(2)For the weight matrix of the second class sample, b(1)For the first kind
The bias matrix of sample, b(2)For the bias matrix of the second class sample, x is sample data, SA、SBRespectively indicate the mould of every class sample
Paste degree of membership, e1、e2It is the vector that element value is 1, ξ and η are slack variable, ξ > 1, η > 1, c1、c2The class that is positive respectively and negative
The penalty factor of class, c1> 0, c2> 0；
Wherein, " one-to-one " classification policy log data inputs in classifier respectively, and each classifier is to a kind of lithology class
It is not predicted, acquired results use " temporal voting strategy ": if the judgement of classifier 1 show that some lithology test data belongs to lithology 1,
Then the poll of lithology 1 adds 1, finally counts each classification number of votes obtained, and test data final result belongs to the most lithology of poll
Classification；
Step 4: model training is carried out on the training set obtained in step 2, and the penalty factor of positive class and negative class is set, according to
According to the optimization method of two hyperplane, best w and b value is found；
Step 5: test model recognition performance on test set in step 2, continuous iteration tests collection, if rear an iteration
Recognition accuracy is lower than preceding an iteration recognition accuracy then return step 4, resets the penalty factor of positive class and negative class, if
An iteration recognition accuracy is higher than preceding an iteration recognition accuracy afterwards, then the punishment of the positive class and negative class of record this time setting
Factor value, when front and back iteration recognition accuracy error stablize within 0.001, then stop iteration.
Beneficial outcomes of the invention are:
(1) more twin support vector machines Lithology Identification Methods of classification are improved not need compared with the methods of existing cross plot
The artificial selection for carrying out log, but after entering data into, algorithm realizes the comprehensive analysis of log automatically, avoids
The uncertain influence to Lithology Discrimination accuracy rate of complicated manual features selection.
(2) improve more twin support vector cassification algorithms of classification can be improved compared with existing traditional classification algorithm
The few lithology classification accuracy rate of sample number makes general lithology predictablity rate more rationally, accurately.
Detailed description of the invention
Fig. 1 is that the present invention is based on the Lithology Discrimination flow charts for improving more twin support vector machines of classification, including data to locate in advance
Reason, data set divide, find optimal solution and verifying collection verifying etc..
Fig. 2 is " one-to-one " more classification policy method schematic diagrams, explains two classification problems being extended to more classification problems
Concrete methods of realizing, log data is inputted in classifier respectively, each classifier predicts a kind of lithology classification, system
Each classification number of votes obtained is counted, test data final result belongs to the most lithology classification of poll.
Specific embodiment
The present invention is described in detail below in conjunction with attached drawing.Fig. 1 is overall flow figure of the invention.Tool of the invention
Body implementation steps are as follows:
Step 1: by data class sample over-sampling small in data set T, obtaining the new data set T' of all kinds of balances, specifically adopt
Quadrat method are as follows:
(1) assume that original data set T there are two class data, respectively P={ p1,p2..., pi,…,pn, N={ n1,n2,…,
ni,…,nm, and n < m；
(2) P={ p is found out1,p2..., pi,…,pnCentral point P ' in set.Calculate separately P ' and { p1,p2...,
pi..., pn } in each data point Euclidean distance, obtain P ' with apart from K nearest data point；
(3) s sample is selected to be denoted as { p from K neighbour's sample1,p2... ps, and s < K < n, respectively in P ' and { p1,
p2... psBetween carry out linear interpolation.Linear interpolation rule are as follows: synthesize new sample Sj=pi+rj×dj, j=1,2 ..., s,
Middle djFor P ' and { p1,p2... psBetween distance, r indicate 0 and 1 between random number；
(4) by the new data point S of synthesisjIt is added to original data set P={ p1,p2..., pnIn, form new training sample
T′；
Step 2: the log data that step 1 is obtained pre-processes, and makes label.
Normalize formula are as follows:
In formula, xiI-th of sample before indicating normalization, xi' indicate i-th of sample after normalization, ximinIndicate data
Concentrate minimum value, ximaxIndicate maximum value in data set.
Log data lithology label is converted into one-hot coding form, such as: lithology classification totally 9 class, marine facies siltstone page
Rock is the 4th class, and original tag 3, one-hot coding form label is [0,0,0,1,0,0,0,0].
Step 3: design a kind of new fuzzy membership function:
If matrixWithThe data sample of positive class and negative class is respectively indicated, wherein matrix A and B's is every
A line respectively indicates a data sample, m1Indicate the data sample number of positive class, m2Indicate that the sample number of negative class, n are data sample
Dimension, the center of positive classThe center of negative classDistance d of the positive class each point away from its central pointA=| | Ai-CA|
|, distance d of the negative class each point away from its central pointB=| | Bi-CB| |, positive class radius rA=max | Ai-CA|, negative class radius rB=max |
Bi-CB|。
For i, i=1,2 each in data set ..., m1+m2, the fuzzy membership S that is proposediIt is as follows:
δ=10 in formula-5。
Fig. 2 is " one-to-one " classification policy schematic diagram, method particularly includes:
(1) more twin support vector machine classifiers of classification will be improved to be arranged between any two classes lithology sample, 9 classifications
Sample need classifier number be 36；
(2) when being tested, test set data are inputted respectively in the more twin support vector machine classifiers of classification of improvement, often
A classifier predicts a kind of lithology classification；
(3) acquired results use " temporal voting strategy ": if the judgement of classifier 1 show that some lithology test data belongs to lithology 1,
Then the poll of lithology 1 adds 1, finally counts each classification number of votes obtained, and test data final result belongs to the most lithology of poll
Classification.
Step 4: model training is carried out on the training set obtained in step 2, and the penalty factor of positive class and negative class is set, according to
According to the optimization method of two hyperplane, best w and b value is found；
Wherein, the optimization method of two hyperplane is as follows:
In formula, SA、SBRespectively indicate the fuzzy membership of every class sample, e1、e2It is the vector that element value is 1, ξ and η are
Slack variable, ξ > 1, η > 1, c1、c2The penalty factor of the class that is positive respectively and negative class, c1> 0, c2> 0；Formula (1) and formula (2)
Dual problem is respectively as follows:
Wherein, H=[A, e1], G=[B, e2], a and γ are Lagrange multiplier vectors.
When data set is Nonlinear separability.Introduce nuclear matrixRequired classifying face
For K (xT,CT)u(1)+b(1)=0 and K (xT,CT)u(2)+b(2)=0, C=[A, B]T, K (X, X ') is kernel function.Optimization problem is
Enable S=[K (A, CT),e1], R=[K (B, CT),e2], dual problem are as follows:
Step 5: test model recognition performance on test set in step 2, continuous iteration tests collection, if rear an iteration
Recognition accuracy is lower than preceding an iteration recognition accuracy then return step 4, resets the penalty factor of positive class and negative class, if
An iteration recognition accuracy is higher than preceding an iteration recognition accuracy afterwards, then the punishment of the positive class and negative class of record this time setting
Factor value, when front and back iteration recognition accuracy error stablize within 0.001, then stop iteration.
Claims (1)
1. a kind of based on the Recognition of Weil Logging Lithology method for improving more twin support vector machines of classification, it is characterised in that including following step
It is rapid:
Step 1: by data class sample over-sampling small in data set T, obtaining the new data set T' of all kinds of balances；
Step 2: new data set T' is pre-processed, and makes label, indicates the label of k class lithology with 0~k-1, k be greater than
New log data collection is divided into training set and test set by the positive integer equal to 1；
Step 3: designing a kind of new fuzzy membership function, construction obscures two Optimal Separating Hyperplanes of twin support vector machines
Optimization method, by " one-to-one " more classification policies and obscuring in conjunction with twin support vector machines using new fuzzy membership function；
If matrixWithThe data sample of positive class and negative class is respectively indicated, wherein every a line of matrix A and B
Respectively indicate a data sample, m1Indicate the data sample number of positive class, m2Indicate that the sample number of negative class, n are data sample dimension
Degree, the center of positive classThe center of negative classDistance d of the positive class each point away from its central pointA=| | Ai-CA| |,
Distance d of the negative class each point away from its central pointB=| | Bi-CB| |, positive class radius rA=max | Ai-CA|, negative class radius rB=max | Bi-
CB|, AiFor a certain sample, B in AiFor a certain sample in B；
For i, i=1,2 ... each in data set, m1+m2, the fuzzy membership S that is proposediIt is as follows:
δ=10 in formula-5；
Wherein, the optimization method of two hyperplane is as follows:
In formula, w(1)For the weight matrix of first kind sample, w(2)For the weight matrix of the second class sample, b(1)For first kind sample
Bias matrix, b(2)For the bias matrix of the second class sample, x is sample data, SA、SBRespectively indicate the fuzzy person in servitude of every class sample
Category degree, e1、e2It is the vector that element value is 1, ξ and η are slack variable, ξ > 1, η > 1, c1、c2The class that is positive respectively and negative class
Penalty factor, c1> 0, c2> 0；
Wherein, " one-to-one " classification policy log data inputs in classifier respectively, each classifier to a kind of lithology classification into
Row prediction, acquired results use " temporal voting strategy ": if the judgement of classifier 1 show that some lithology test data belongs to lithology 1, rock
The poll of property 1 adds 1, finally counts each classification number of votes obtained, test data final result belongs to the most lithology classification of poll；
Step 4: carrying out model training on the training set obtained in step 2, the penalty factor of positive class and negative class is set, according to two
The optimization method of a hyperplane finds best w and b value；
Step 5: test model recognition performance on test set in step 2, continuous iteration tests collection, if rear an iteration identification
Accuracy rate is lower than preceding an iteration recognition accuracy then return step 4, the penalty factor of positive class and negative class is reset, if latter
Secondary iteration recognition accuracy is higher than preceding an iteration recognition accuracy, then the penalty factor of the positive class and negative class of record this time setting
Numerical value, when front and back iteration recognition accuracy error stablize within 0.001, then stop iteration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910781790.4A CN110516733A (en) | 2019-08-23 | 2019-08-23 | A kind of Recognition of Weil Logging Lithology method based on the more twin support vector machines of classification of improvement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910781790.4A CN110516733A (en) | 2019-08-23 | 2019-08-23 | A kind of Recognition of Weil Logging Lithology method based on the more twin support vector machines of classification of improvement |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110516733A true CN110516733A (en) | 2019-11-29 |
Family
ID=68626278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910781790.4A Pending CN110516733A (en) | 2019-08-23 | 2019-08-23 | A kind of Recognition of Weil Logging Lithology method based on the more twin support vector machines of classification of improvement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110516733A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111639722A (en) * | 2020-06-22 | 2020-09-08 | 杭州电力设备制造有限公司 | Transformer fault diagnosis method based on principal component analysis and twin support vector machine |
| CN111654874A (en) * | 2020-06-03 | 2020-09-11 | 枣庄学院 | Wireless sensor network anomaly detection method |
| CN111767803A (en) * | 2020-06-08 | 2020-10-13 | 北京理工大学 | Identification method for anti-target attitude sensitivity of synthetic extremely-narrow pulse radar |
| CN112991370A (en) * | 2021-04-08 | 2021-06-18 | 重庆大学 | Rock core CT crack identification and segmentation method |
| CN113780346A (en) * | 2021-08-06 | 2021-12-10 | 中国科学技术大学 | Method and system for adjusting prior constraint classifier and readable storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102071928A (en) * | 2009-11-25 | 2011-05-25 | 中国石油天然气股份有限公司 | Method for identifying lithology of three-dimensional volcanic rock |
| CN103745234A (en) * | 2014-01-23 | 2014-04-23 | 东北大学 | Band steel surface defect feature extraction and classification method |
| CN105388531A (en) * | 2015-10-19 | 2016-03-09 | 成都理工大学 | Lithology identification method based on support vector regression machine and Kernel Fisher discriminant analysis |
| US20160124099A1 (en) * | 2014-11-05 | 2016-05-05 | Petrochina Company Limited | Method and system for identification of geology lithological difference |
| CN106295153A (en) * | 2016-08-03 | 2017-01-04 | 南京航空航天大学 | A kind of Fault Diagnosis of Aircraft Engine Gas Path method based on twin support vector machine |
| CN107292350A (en) * | 2017-08-04 | 2017-10-24 | 电子科技大学 | The method for detecting abnormality of large-scale data |
| CN108171365A (en) * | 2017-12-15 | 2018-06-15 | 南京理工大学 | A kind of traffic status prediction method based on improvement SVM algorithm |
| CN109165694A (en) * | 2018-09-12 | 2019-01-08 | 太原理工大学 | The classification method and system of a kind of pair of non-equilibrium data collection |
| CN109919184A (en) * | 2019-01-28 | 2019-06-21 | 中国石油大学(北京) | A kind of more well complex lithology intelligent identification Methods and system based on log data |
| CN109934280A (en) * | 2019-03-07 | 2019-06-25 | 贵州大学 | A kind of unbalanced data classification method based on PSO-DEC-IFSVM sorting algorithm |
-
2019
- 2019-08-23 CN CN201910781790.4A patent/CN110516733A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102071928A (en) * | 2009-11-25 | 2011-05-25 | 中国石油天然气股份有限公司 | Method for identifying lithology of three-dimensional volcanic rock |
| CN103745234A (en) * | 2014-01-23 | 2014-04-23 | 东北大学 | Band steel surface defect feature extraction and classification method |
| US20160124099A1 (en) * | 2014-11-05 | 2016-05-05 | Petrochina Company Limited | Method and system for identification of geology lithological difference |
| CN105388531A (en) * | 2015-10-19 | 2016-03-09 | 成都理工大学 | Lithology identification method based on support vector regression machine and Kernel Fisher discriminant analysis |
| CN106295153A (en) * | 2016-08-03 | 2017-01-04 | 南京航空航天大学 | A kind of Fault Diagnosis of Aircraft Engine Gas Path method based on twin support vector machine |
| CN107292350A (en) * | 2017-08-04 | 2017-10-24 | 电子科技大学 | The method for detecting abnormality of large-scale data |
| CN108171365A (en) * | 2017-12-15 | 2018-06-15 | 南京理工大学 | A kind of traffic status prediction method based on improvement SVM algorithm |
| CN109165694A (en) * | 2018-09-12 | 2019-01-08 | 太原理工大学 | The classification method and system of a kind of pair of non-equilibrium data collection |
| CN109919184A (en) * | 2019-01-28 | 2019-06-21 | 中国石油大学(北京) | A kind of more well complex lithology intelligent identification Methods and system based on log data |
| CN109934280A (en) * | 2019-03-07 | 2019-06-25 | 贵州大学 | A kind of unbalanced data classification method based on PSO-DEC-IFSVM sorting algorithm |
Non-Patent Citations (4)
| Title |
|---|
| 徐亚军: "中子寿命测井层位自动识别的研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
| 李凯等: "引入调整项的模糊孪生支持向量机", 《计算机科学》 * |
| 牟丹等: "基于ＳＶＭ 测井数据的火山岩岩性识别———以辽河盆地东部坳陷为例", 《地球物理学报》 * |
| 程元启等: "基于模糊支持向量机的软件缺陷预测技术", 《计算机工程与设计》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111654874A (en) * | 2020-06-03 | 2020-09-11 | 枣庄学院 | Wireless sensor network anomaly detection method |
| CN111654874B (en) * | 2020-06-03 | 2023-02-24 | 枣庄学院 | Wireless sensor network anomaly detection method |
| CN111767803A (en) * | 2020-06-08 | 2020-10-13 | 北京理工大学 | Identification method for anti-target attitude sensitivity of synthetic extremely-narrow pulse radar |
| CN111767803B (en) * | 2020-06-08 | 2022-02-08 | 北京理工大学 | Identification method for anti-target attitude sensitivity of synthetic extremely-narrow pulse radar |
| CN111639722A (en) * | 2020-06-22 | 2020-09-08 | 杭州电力设备制造有限公司 | Transformer fault diagnosis method based on principal component analysis and twin support vector machine |
| CN112991370A (en) * | 2021-04-08 | 2021-06-18 | 重庆大学 | Rock core CT crack identification and segmentation method |
| CN112991370B (en) * | 2021-04-08 | 2022-11-25 | 重庆大学 | Rock core CT crack identification and segmentation method |
| CN113780346A (en) * | 2021-08-06 | 2021-12-10 | 中国科学技术大学 | Method and system for adjusting prior constraint classifier and readable storage medium |
| CN113780346B (en) * | 2021-08-06 | 2023-06-16 | 中国科学技术大学 | Priori constraint classifier adjustment method, system and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110516733A (en) | A kind of Recognition of Weil Logging Lithology method based on the more twin support vector machines of classification of improvement | |
| CN107194433B (en) | A kind of Radar range profile's target identification method based on depth autoencoder network | |
| CN102521656B (en) | Integrated transfer learning method for classification of unbalance samples | |
| CN108520357B (en) | Method and device for judging line loss abnormality reason and server | |
| CN103632168B (en) | Classifier integration method for machine learning | |
| CN109034194B (en) | Transaction fraud behavior deep detection method based on feature differentiation | |
| CN103489005B (en) | A kind of Classification of High Resolution Satellite Images method based on multiple Classifiers Combination | |
| CN109765333A (en) | A kind of Diagnosis Method of Transformer Faults based on GoogleNet model | |
| CN107103332A (en) | A kind of Method Using Relevance Vector Machine sorting technique towards large-scale dataset | |
| CN110298321A (en) | Route denial information extraction based on deep learning image classification | |
| CN106845717A (en) | A kind of energy efficiency evaluation method based on multi-model convergence strategy | |
| CN109711429A (en) | A kind of evaluating reservoir classification method and device | |
| CN107944472B (en) | A kind of airspace operation situation calculation method based on transfer learning | |
| CN106599913A (en) | Cluster-based multi-label imbalance biomedical data classification method | |
| CN108363810A (en) | A kind of file classification method and device | |
| CN109214460A (en) | Method for diagnosing fault of power transformer based on Relative Transformation Yu nuclear entropy constituent analysis | |
| CN110348608A (en) | A kind of prediction technique for improving LSTM based on fuzzy clustering algorithm | |
| CN105653450A (en) | Software defect data feature selection method based on combination of modified genetic algorithm and Adaboost | |
| CN105005789B (en) | A kind of remote sensing images terrain classification method of view-based access control model vocabulary | |
| CN112039687A (en) | Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network | |
| CN106066493A (en) | Bayes's petrofacies method of discrimination and device | |
| CN110245252A (en) | Machine learning model automatic generation method based on genetic algorithm | |
| CN104427505B (en) | A kind of method and device of cell scenario division | |
| CN106156401A (en) | Data-driven system state model on-line identification methods based on many assembled classifiers | |
| CN107644057A (en) | A kind of absolute uneven file classification method based on transfer learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191129 |