CN110427957A - A kind of classification method and device of the geochemistry data based on machine learning - Google Patents
A kind of classification method and device of the geochemistry data based on machine learning Download PDFInfo
- Publication number
- CN110427957A CN110427957A CN201910507428.8A CN201910507428A CN110427957A CN 110427957 A CN110427957 A CN 110427957A CN 201910507428 A CN201910507428 A CN 201910507428A CN 110427957 A CN110427957 A CN 110427957A
- Authority
- CN
- China
- Prior art keywords
- data
- classifier
- geochemistry
- machine learning
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
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
Abstract
The embodiment of the present invention discloses the classification method and device of a kind of geochemistry data based on machine learning.The method, comprising: step 1, obtain trained geochemistry data；Step 2, Supervised machine learning is carried out with geochemistry data to the training, generates classifier；Step 3, the geochemistry data to be sorted that will be collected into, is classified using the classifier.
Description
Technical field
The present invention relates to Geology research and geologic prospect field more particularly to a kind of earth based on machine learning
The classification method and device of chemical data.
Background technique
The constituent analysis of geological sample is the important research method of geoscience research, for it is deep understand earth evolution,
Mineral resources formation and mineral exploration etc. have a very important significance.It is Main elements based on geological sample, micro
Element, isotopic composition analyze data, classify to geological sample, can be used for determining origin of ore deposit, rock type, diagenesis at
Mine structural environment, delineation target prospecting area etc..
By current geochemical analysis means, the multiple element of same geological sample can be obtained, isotope contains
Amount.As Within Monominerals in-situ micro area constituent analysis (use the methods of LA-ICP-MS, SIMS, SHRIMP, electron probe), total rock or
The data such as composition of ores analysis (using the methods of XRF, ICP-MS, Fire Assaying), geochemical sample constituent analysis, can all obtain simultaneously
A variety of Main elements of same sample, microelement, isotopic composition.These geochemistry datas mathematically show as multidimensional
Data, this to carry out data Accurate classification and the difficulty of excavation increases.
At present to the classification of geochemistry data mainly using a series of two-dimentional disciminating diagrams (such as Pearce diagram).This
Though a little disciminating diagrams to geochemical investigation play the role of it is certain actively promote, usual classification accuracy is not high, difference
The geological sample of the origin cause of formation is usually present the overlapping region of large area after two-dimentional throwing figure.Trace it to its cause is due to geochemistry data
For high dimensional data, compressing it will cause significantly information loss in two dimension diagram, therefore not using two dimension diagram
It can be carried out effective differentiation.Therefore, it is necessary to explore a kind of method that directly can be classified to high dimensional data and be handled.
Machine learning is the multi-field intersection sexology developed in recent years, in the field such as SVM (supporting vector
Machine), neural network and decision tree scheduling algorithm can effectively classify to high dimensional data.Therefore inventor utilizes machine learning skill
Art proposes a kind of new geochemistry data classification processing method, and is tested by feasibility of the specific example to this method
Card finally thinks that this method can carry out Accurate classification to higher-dimension geochemistry data and depth is excavated.
Summary of the invention
In view of this, the embodiment of the present invention provides the classification method and dress of a kind of geochemistry data based on machine learning
It sets, can classify to higher-dimension geochemistry data.
A kind of classification method of the geochemistry data based on machine learning, comprising:
Step 1, trained geochemistry data is obtained；
Step 2, Supervised machine learning is carried out with geochemistry data to the training, generates classifier；
Step 3, the geochemistry data to be sorted that will be collected into, is classified using the classifier.
A kind of sorter of the geochemistry data based on machine learning, comprising:
Acquiring unit obtains trained geochemistry data；
Unit carries out Supervised machine learning with geochemistry data to the training, generates classifier,
Taxon, the geochemistry data to be sorted that will be collected into are classified using institute's classifier.
Machine learning method is applied to such as geological sample Main elements, microelement, isotopic composition number by the present invention
According to equal analysis and interpretation, can classify to higher-dimension geochemistry data.
Detailed description of the invention
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 technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the schematic diagram of the classification method of geochemistry data of the embodiment of the present invention based on machine learning；
Fig. 2 is the schematic diagram of the sorter of the geochemistry data of the invention based on machine learning；
Fig. 3 is that pyrite microelement two dimension throws figure in application scenarios of the present invention；
Fig. 4 is that pyrite microelement three-dimensional throws figure in application scenarios of the present invention；
Fig. 5 is that pyrite microelement higher-dimension throws figure in application scenarios of the present invention；
Fig. 6 is respectively to beg ore-pyrite microelement classification results figure suddenly in application scenarios of the present invention；
Fig. 7 is the raw disk ore-pyrite microelement classification results figure of first in application scenarios of the present invention.
Specific embodiment
The embodiment of the present invention is described in detail with reference to the accompanying drawing.
It will be appreciated that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts it is all its
Its embodiment, shall fall within the protection scope of the present invention.
For convenience of description, description apparatus above is to be divided into various units/modules with function to describe respectively.Certainly, In
Implement to realize each unit/module function in the same or multiple software and or hardware when the present invention.
As shown in Figure 1, being a kind of classification method of the geochemistry data based on machine learning of the present invention, packet
It includes:
Step 1, trained geochemistry data is obtained；For example, the step 1 includes:
Obtain VMS VHMS-type deposit, Orogenic-Type mineral deposit, appositional pattern pyrite original position LA-ICP-MS microelement
Data form training set, as trained geochemistry data.
Step 11, the geochemistry data is cleaned.The step 11 includes: to find out instruction using statistical algorithms
Experienced wrong data and repeated data in geochemistry data is cleaned, and the data value of vacancy is with the detection limit generation of equipment
It replaces.
Step 2, Supervised machine learning is carried out with geochemistry data to the training, generates classifier；
Step 21, leave-one-out leaving-one method cross detection is carried out to the classifier.
The step 21 includes:
By in training set in n sample value as performance number be isolated, as unknown data, using remaining n-1 data into
Row supervised learning obtains classifier in turn；
Classified using classifier to the sample for being taken as unknown data, by classification results and the sample actual result into
Row compares, and determines whether classifier correctly classifies；
The cross-beta carries out n times, obtain n classify as a result, and finally obtaining the accuracy of classifier.
For example, the step 21 includes:
The pyrite in VMS VHMS-type deposit, Orogenic-Type mineral deposit and sedimentary formation is obtained, amounts to n
Mineral deposit or stratigraphic unit；
An optional mineral deposit is as unknown sample, and using the data in remaining n-1 mineral deposit as training set, training can distinguish three
The classifier of kind pyrite genetic type；
Classified using the classifier that training obtains to the sample selected, by acquired results and the practical mineral deposit class of the sample
Type comparison, it is determined whether correct.
Step 3, the geochemistry data to be sorted that will be collected into, is classified using the classifier.
As shown in Fig. 2, being a kind of sorter of the geochemistry data based on machine learning of the present invention, packet
It includes:
Acquiring unit 21 obtains trained geochemistry data；
Unit 22 carries out Supervised machine learning with geochemistry data to the training, generates classifier,
Taxon 23, the geochemistry data to be sorted that will be collected into are classified using institute's classifier.
Application scenarios of the invention are described below.
Machine learning method is applied to geological sample Main elements, microelement, isotopic composition data by the present invention
Analysis and interpretation, can classify to higher-dimension geochemistry data.
The invention be it is a kind of suitable for geochemistry high dimensional data classification technical method, including 3 aspect: data cleansing and
The processing of vacancy value trains classifier using Supervised machine learning method, tests the validity of classifier.
(1) Supervised machine learning method training classifier is utilized
Supervised learning is a kind of study form of machine learning, is mainly used in classification and prediction.In this process, defeated
Entering data and is referred to as " training data ", every group of training data has a specific actual result, when establishing prediction model,
A learning process is established in supervised study, and prediction result is compared with the actual result of " training data ", continuous to adjust
Whole prediction model, until the prediction result of model reaches an expected accuracy rate.Through practicing, following sorting algorithm is to geochemical
The classifying quality for learning data is preferable:
SVM (Support Vector Mach ine), i.e. support vector machines are a kind of for two classification and multi-class to ask
Inscribe the model of classification.Main thought is the hyperplane that can scratch all data samples found in space, and is made
The distance for obtaining all data to this hyperplane in sample set is most short.Find that nonlinear support vector machines can be effective through practice
Complete the classification of geochemistry high dimensional data.
Artificial neural network (ANN), artificial neural network are a kind of algorithm structures, and machine is enabled to learn all.Through
Practice discovery, the neural network containing one or two hidden layer can construct the geochemistry data classifier of high accuracy.
Decision tree, also known as decision tree are a kind of tree constructions for applying to classification, each internal node representative pair therein
The primary test of a certain attribute, each edge represent a test result, and leaf node represents the distribution of some class or class.Decision tree
Decision process is needed since the root node of decision tree, and testing data is compared with the characteristic node in decision tree, and according to
The next relatively branch of comparison result selection selection, until leaf node is as the final result of decision.It is found through practice, decision tree
Geochemistry data classification problem, and the advantage that can be explained with algorithm can effectively be solved.
(2) data cleansing
In obtained sample data, usually there is missing data, it is unfavorable that this can generate subsequent machine-learning process
It influences.The content of the element or isotope is lower than instrument in the reason of causing missing values in geochemistry data mainly sample
Detection limit.Therefore, practice discovery, directly substitutes missing values using detection limit, can obtain good classifying quality.In addition, will
All geochemical composition datas take logarithm, can obtain preferable Gaussian Profile, be conducive to machine learning.
(3) it tests to the validity of classifier
It is found through practice, the maximally efficient method of inspection is leave-one-out cross-beta.I.e. by n in training set
A certain sample value isolation in sample value carries out supervised learning using remaining n-1 data and then obtains as unknown data
Classifier.Classified using classifier to the sample for being taken as unknown data, by classification results and the sample actual result into
Row compares, and determines whether classifier correctly classifies.The cross-beta can carry out n times, obtain n classification as a result, simultaneously final
To the accuracy of classifier.
Using method proposed by the invention, classifier is established using to pyrite micronutrient levels, it can be with effective district
Divide deposit type.The study on the genesis that the classifier is respectively begged to mineral deposit applied to the raw disk of first, suddenly, can effectively differentiate the complexity of the mine
At mine history.Inventor is collected into about 2000 pyrite original position LA-ICP-MS trace element datas, these data from
Pyrite in VMS type mineral deposit, Orogenic-Type mineral deposit and sedimentary formation amounts to 99 mineral deposits or stratigraphic unit, using known to these
The pyrite of the origin cause of formation carries out Supervised machine learning as training set, and utilizes leave-one-out cross-beta inspection-classification
The classification capacity of device, the accuracy for finally obtaining classifier is 99%, and showing can be to higher-dimension geochemistry using machine learning
Data carry out highly effective classification.The study on the genesis that it is applied to the raw disk of first, respectively begs mineral deposit suddenly by the classification, is disclosed to achievement
The complexity of the mine is at mine history.
It is carried out below with being applied to such as northern China Lang shan Mountain-slag that Mount Taishan metallogenic belt (research area) interior origin of ore deposit research
Explanation.
Embodiment: the Lang shan Mountain-slag that two typical ore deposits origin cause of formation of Mount Taishan metallogenic belt judgement
The present invention successively the following steps are included:
Step 1, VMS, Orogenic-Type, appositional pattern pyrite original position LA-ICP-MS trace element data are obtained, it is clear to carry out data
It washes, the processing of vacancy value and etc., and compare two dimension, three-dimensional and multidimensional data classification accuracy rate.
Data, which are all derived from, publishes data from the University of Tasmania laboratory CODES, including VMS block
The pyrite LA-ICP-MS of totally 99 practical mineral deposits or sedimentary formation is micro for shape sulfide deposit, Orogenic-Type and appositional pattern three classes
Element data more than totally 2000, as training set.Wrong data and repeated data therein is found out using statistical algorithms to carry out
The data value of cleaning, vacancy is replaced with the detection limit of equipment.
For the relationship for illustrating different dimensions and classification accuracy, two dimension, three-dimensional and hyperspace are carried out to data first
Figure is thrown, as a result as follows: Fig. 3 is that pyrite microelement two dimension throws figure；Fig. 4 is that pyrite microelement three-dimensional throws figure；Fig. 5 is Huang
Iron ore microelement higher-dimension throws figure.
By Fig. 3 and Fig. 4 it is found that no matter the pyrite micronutrient levels data of higher-dimension throw in two dimension or three-dimensional space
There is the overlapping region in three types mineral deposit in figure, is unable to judge accurately deposit type in the area.Statistics shows that two dimension is sentenced
The classification accuracy of not figure is only 66.2%, and three-dimensional differentiates that figure classification accuracy is 70% or so.And when dimension increases, point
Class accuracy rate also increases, and reaches as high as 99.8%, shows that the throwing figure of higher dimensional space can distinguish different origin of ore deposit well
Pyrite.
Step 2, Supervised machine learning, training classifier are carried out to sample data, and carries out leave-one-out friendship
Fork detection.
Using the compositional data of this more than 2000 samples of pyrite as training set, with pyrite totally 14 kinds of micronutrient levels
As input value, using its origin cause of formation as target value, using the learning method (herein using SVM method) for having supervision, training is obtained
Classifier.
Utilize the accuracy of leave-one-out cross-beta inspection-classification device.An optional mineral deposit as unknown sample,
Using the data in remaining 98 mineral deposits as training set, training can distinguish the classifier of three kinds of pyrite genetic types, utilize training
Obtained classifier classifies to the sample selected, and acquired results and the practical deposit type of the sample are compared, it is determined whether
Correctly.The cross-beta carries out 99 times altogether, obtains 98 correct classification, and final classification device accuracy is 99%, shows to utilize machine
Device study can be used for origin of ore deposit research to the method that pyrite microelement is classified.
Step 3, it is applied to practical deposit classification
Step 3.1, the classification to mineral deposit is respectively begged suddenly.
Respectively begging suddenly is a large size Cu-Pb-Zn polymetallic deposit in your Mount Taishan metallogenic belt of the Lang shan Mountain-slag, quite at the mine origin cause of formation
It is disputed on, groups of people think to be formed in Proterozoic Era jet flow deposition, and some scholars are research shows that it is the rear heat by construction control
Liquid mineral deposit.The classifier of be collected into 45 pyrite trace element data training is classified, as a result as follows: Fig. 6 is
Ore-pyrite microelement classification results figure is respectively begged suddenly
As shown, classification results show that the ore-pyrite Trace Elements Features meet Orogenic-Type Ore Deposit Features, accurately
Rate is 94%, show respectively to beg suddenly formation of ore deposits in the later period with orogenetic hydrothermal stage the non-deposited phase.In addition, other rock chalcographys,
Fluid inclusion, Isotope Research and the research in mineralogenetic epoch demonstrate this point.
Step 3.2, to the classification in the raw disk mineral deposit of first
The raw disk mineral deposit of first is similar with mineral deposit is respectively begged suddenly, is also a large size Zn-Pb polymetallic deposit in the metallogenic belt.It is right
Its origin cause of formation equally exists two kinds of viewpoints similar with mineral deposit is respectively begged suddenly.Microexamination is found
Particulate bulk pyrite, advanced stage is coarse grain hydrothermal solution veiny pyrite, and the main equal output of Zn-Pb mineral is in advanced stage.Utilize training
Classifier classify to two phase pyrite microelements, classification results are as shown in Figure 7.Classification results show that particulate is blocky yellow
Iron ore is the Union dyeing origin cause of formation, and coarse-grained pyrite is Orogenic-Type origin of ore deposit, in conjunction with microscopic findings, shows the raw Pan Kuang of first
For bed there are two phase mineralisings, early stage deposits major developmental pyritization, and advanced stage is formed the Zn-Pb mineralising of Orogenic-Type by construction control.
The mode is also proved by a series of researchs such as chalcography, fluid inclusion, Thermodynamic Simulation.Fig. 7 is the raw disk ore-pyrite of first
Microelement classification results figure.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the program can be stored in a computer-readable storage medium
In, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) or random access memory (Random Access
Memory, RAM) etc..
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by those familiar with the art, all answers
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (8)
1. a kind of classification method of the geochemistry data based on machine learning characterized by comprising
Step 1, trained geochemistry data is obtained；
Step 2, Supervised machine learning is carried out with geochemistry data to the training, generates classifier；
Step 3, the geochemistry data to be sorted that will be collected into, is classified using the classifier.
2. the method according to claim 1, wherein after the step 1, before the step 2, the method
Further include:
Step 11, the geochemistry data is cleaned.
3. the method according to claim 1, wherein after the step 2, before the step 3, the method
Further include:
Step 21, leave-one-out leaving-one method cross detection is carried out to the classifier.
4. according to the method described in claim 3, it is characterized in that, the step 1 includes:
Obtain VMS VHMS-type deposit, Orogenic-Type mineral deposit, appositional pattern pyrite original position LA-ICP-MS laser ablation inductance
Coupled plasma mass spectrometry trace element data forms training set, as trained geochemistry data.
5. according to the method described in claim 4, it is characterized in that, the step 21 includes:
By in training set in n sample value as performance number be isolated, as unknown data, had using remaining n-1 data
Supervised learning obtains classifier in turn；
Classified using classifier to the sample for being taken as unknown data, classification results and the sample actual result are compared
It is right, determine whether classifier correctly classifies；
The cross-beta carries out n times, obtain n classify as a result, and finally obtaining the accuracy of classifier.
6. according to the method described in claim 5, it is characterized in that, the step 21 includes:
The pyrite in VMS VHMS-type deposit, Orogenic-Type mineral deposit and sedimentary formation is obtained, n mineral deposit is amounted to
Or stratigraphic unit；
An optional mineral deposit is as unknown sample, and using the data in remaining n-1 mineral deposit as training set, training can distinguish three kinds of Huangs
The classifier of Origin of The Iron Deposits type；
Classified using the classifier that training obtains to the sample selected, by acquired results and the practical deposit type pair of the sample
Than, it is determined whether it is correct.
7. according to the method described in claim 2, it is characterized in that, the step 11 includes:
The wrong data trained in geochemistry data is found out using statistical algorithms and repeated data is cleaned, vacancy
Data value is replaced with the detection limit of equipment.
8. a kind of sorter of the geochemistry data based on machine learning characterized by comprising
Acquiring unit obtains trained geochemistry data；
Unit carries out Supervised machine learning with geochemistry data to the training, generates classifier,
Taxon, the geochemistry data to be sorted that will be collected into are classified using institute's classifier.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910507428.8A CN110427957A (en) | 2019-06-12 | 2019-06-12 | A kind of classification method and device of the geochemistry data based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910507428.8A CN110427957A (en) | 2019-06-12 | 2019-06-12 | A kind of classification method and device of the geochemistry data based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110427957A true CN110427957A (en) | 2019-11-08 |
Family
ID=68407621
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910507428.8A Withdrawn CN110427957A (en) | 2019-06-12 | 2019-06-12 | A kind of classification method and device of the geochemistry data based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110427957A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111967576A (en) * | 2020-07-22 | 2020-11-20 | 长春工程学院 | Geochemical data processing method and system based on deep learning |
| CN113627657A (en) * | 2021-07-23 | 2021-11-09 | 核工业北京地质研究院 | Sandstone-type uranium mineralization interest area prediction method using machine learning model |
| CN114203268A (en) * | 2021-12-07 | 2022-03-18 | 昆明理工大学 | Machine learning-based geochemical data classification method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050197980A1 (en) * | 2004-02-06 | 2005-09-08 | Siemens Medical Solutions Usa, Inc. | System and method for a sparse kernel expansion for a bayes classifier |
| CN106355011A (en) * | 2016-08-30 | 2017-01-25 | 有色金属矿产地质调查中心 | Geochemical data element sequence structure analysis method and device |
| CN107844804A (en) * | 2017-11-10 | 2018-03-27 | 上海杉达学院 | A kind of computational methods of flesh texture MRI and the Data Modeling Method based on it |
-
2019
- 2019-06-12 CN CN201910507428.8A patent/CN110427957A/en not_active Withdrawn
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050197980A1 (en) * | 2004-02-06 | 2005-09-08 | Siemens Medical Solutions Usa, Inc. | System and method for a sparse kernel expansion for a bayes classifier |
| CN106355011A (en) * | 2016-08-30 | 2017-01-25 | 有色金属矿产地质调查中心 | Geochemical data element sequence structure analysis method and device |
| CN107844804A (en) * | 2017-11-10 | 2018-03-27 | 上海杉达学院 | A kind of computational methods of flesh texture MRI and the Data Modeling Method based on it |
Non-Patent Citations (2)
| Title |
|---|
| 焦守涛等: "基于 GEOROC 数据库的全球辉长岩大数据的大地构造环境智能判别研究", 《岩石学报》 * |
| 蔡立志等: "《大数据测评》", 31 January 2015, 上海科学技术出版社 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111967576A (en) * | 2020-07-22 | 2020-11-20 | 长春工程学院 | Geochemical data processing method and system based on deep learning |
| CN111967576B (en) * | 2020-07-22 | 2022-09-02 | 长春工程学院 | Geochemical data processing method and system based on deep learning |
| CN113627657A (en) * | 2021-07-23 | 2021-11-09 | 核工业北京地质研究院 | Sandstone-type uranium mineralization interest area prediction method using machine learning model |
| CN114203268A (en) * | 2021-12-07 | 2022-03-18 | 昆明理工大学 | Machine learning-based geochemical data classification method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106407649B (en) | Microseismic signals based on time recurrent neural network then automatic pick method | |
| CN110442666B (en) | Mineral resource prediction method and system based on neural network model | |
| CN106355011B (en) | Geochemical data element sequence structure analysis method and device | |
| Bigdeli et al. | Application of self-organizing map (SOM) and K-means clustering algorithms for portraying geochemical anomaly patterns in Moalleman district, NE Iran | |
| CN110427957A (en) | A kind of classification method and device of the geochemistry data based on machine learning | |
| Hao et al. | Machine learning application to automatically classify heavy minerals in river sand by using SEM/EDS data | |
| Koch et al. | Automated drill core mineralogical characterization method for texture classification and modal mineralogy estimation for geometallurgy | |
| Li et al. | Prospectivity mapping for tungsten polymetallic mineral resources, Nanling metallogenic belt, south China: Use of random forest algorithm from a perspective of data imbalance | |
| CN106066493A (en) | Bayes's petrofacies method of discrimination and device | |
| Sepúlveda et al. | Fuzzy clustering with spatial correction and its application to geometallurgical domaining | |
| CN109543720A (en) | A kind of wafer figure defect mode recognition methods generating network based on confrontation | |
| CN110097069A (en) | A kind of support vector machines Lithofacies Identification method and device based on depth Multiple Kernel Learning | |
| Nykänen et al. | Spatial analysis techniques as successful mineral-potential mapping tools for orogenic gold deposits in the northern Fennoscandian Shield, Finland | |
| Ding et al. | Siamese network based prospecting prediction method: A case study from the Au deposit in the Chongli mineral concentrate area in Zhangjiakou, Hebei Province, China | |
| Chen et al. | Global oceanic eddy identification: A deep learning method from argo profiles and altimetry data | |
| Rammelkamp et al. | Clustering supported classification of ChemCam data from Gale crater, Mars | |
| CN113033637B (en) | Integrated learning lithofacies prediction method based on rock physical sample enhancement | |
| CN106250676A (en) | Element geochemistry survey data method for optimizing based on information gain-ratio | |
| CN116298190A (en) | Screening and determining method for geochemical investigation index elements | |
| Olatunji | Comparison of Extreme Learning Machines and Support Vector Machines on Premium and Regular Gasoline Classification for Arson and Oil Spill Investigation. | |
| Peters et al. | The Southern New England Orogen Mineral Potential Project | |
| Li et al. | Prospectivity and Uncertainty Analysis of Tungsten Polymetallogenic Mineral Resources in the Nanling Metallogenic Belt, South China: A Comparative Study of AdaBoost, GBDT, and XgBoost Algorithms | |
| Abbas et al. | Unsupervised machine learning technique for classifying production zones in unconventional reservoirs | |
| Konaté et al. | Machine learning interpretation of conventional well logs in crystalline rocks | |
| Cevik | Machine Learning Enhancements for Knowledge Discovery in Mineral Exploration and Improved Mineral Resource Classification |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20191108 |
|
| WW01 | Invention patent application withdrawn after publication |