CN109711597A - A kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model - Google Patents
A kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model Download PDFInfo
- Publication number
- CN109711597A CN109711597A CN201811354629.0A CN201811354629A CN109711597A CN 109711597 A CN109711597 A CN 109711597A CN 201811354629 A CN201811354629 A CN 201811354629A CN 109711597 A CN109711597 A CN 109711597A
- Authority
- CN
- China
- Prior art keywords
- random forest
- copper
- forest model
- nickel sulfide
- stratified random
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model that the invention discloses a kind of, comprising the following steps: S1: polynary geologic information in collecting zone establishes Copper-nickel Sulfide Ore Deposit geographc information data base；S2: the Copper-nickel Sulfide Ore Deposit regularity of ore formation in analyzed area extracts ore information；S3: choosing non-mine point, and known mine point is combined to construct training sample set, training stratified random forest model；S4: optimizing stratified random forest model, and carries out metallogenic prognosis using Optimized model；S5: verifying prediction result assesses ore information importance.The present invention carries out Copper-nickel Sulfide Ore Deposit metallogenic prognosis using stratified random forest model, it can effectively solve the problems, such as mine point mistake minute caused by training sample imbalance, predictablity rate decline, to more objective, accurately evaluation this area's Copper-nickel Sulfide Ore Deposit minerogenic potentiality, lay the foundation for next step work of exploration and development.
Description
Technical field
The present invention relates to Copper-nickel Sulfide Ore Deposit minerogenic potentiality assessment technology fields, are based on dividing more particularly, to one kind
The Copper-nickel Sulfide Ore Deposit metallogenic prognosis method of layer Random Forest model.
Background technique
Nickel is a kind of important noble metal, has the characteristics such as high temperature resistant, anti-oxidant, anticorrosive, is widely used in civil, army
The fields such as work, medical treatment have important strategic importance.It is increasingly reduced however as superficial part nickel minerals and nickel minerals easy to identify, nickel minerals
The difficulty of preliminry basic research becomes increasing.Thus, the MINERAL PREDICTION method that utilization is more acurrate, more efficient improves copper-nickel sulphide
Mineral exploration working efficiency seems very necessary.
Using machine learning model carry out metallogenic prognosis can usually obtain it is preferable as a result, because are as follows: (1) it can be accurate
The non-linear space positional relationship of polynary ore information and mine point in region is described；(2) it, which can effectively be integrated, polynary looks for mine to believe
Breath, and the spatial relation according to each ore information and mineral deposit carry out quantitative forecast to minerogenic potentiality in region.
Currently, generalling use balance training collection, that is, included mine point when carrying out metallogenic prognosis using machine learning model
(positive sample) number is equal to the training set of non-mine point (negative sample) number.However in a practical situation, the non-mine point in area is studied
Number is much larger than mine point number, and for conventional machines learning model when handling such uneven sample set, wrong point of meeting even more important
A small amount of sample, i.e. mine point.
Summary of the invention
Present invention aim to address during practical metallogenic prognosis, conventional machines learning model can not utilize uneven sample
The technical issues of this collection is trained, and influences metallogenic prognosis precision proposes a kind of cupro-nickel sulphur based on stratified random forest model
Compound ore deposit prediction technique.
To realize the above goal of the invention, the technical solution adopted is that:
A kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model, comprising the following steps:
S1: polynary geologic information in collecting zone establishes Copper-nickel Sulfide Ore Deposit geographc information data base；
S2: the Copper-nickel Sulfide Ore Deposit regularity of ore formation in analyzed area extracts ore information；
S3: choosing non-mine point, and known mine point is combined to construct training sample set, training stratified random forest model；
S4: optimizing stratified random forest model, and carries out metallogenic prognosis using Optimized model；
S5: verifying prediction result assesses ore information importance.
Preferably, ore information described in step S2 includes control ore-rock magmatic rock information, ore-controlling structure information, geochemistry element
Paragenetic association information.
Further, the geochemical elements paragenetic association information extracting method is Principal Component Analysis, and this method can
More accurately to describe the inner link between different chemical elements.
Preferably, non-mine point selection range described in step S3 is analyzed by spatial point patterns and is determined, it can be ensured that non-mine point
There are notable differences with geologic setting locating for mine point, increase sample reliability.
Preferably, training sample set described in step S3 is non-equilibrium sample set, and the non-mine point number studied in area is long-range
In mine point number.
Non-equilibrium sample set can be effectively treated in stratified random forest model described in step S3, reduce the mistake minute of mine point, mention
High predictablity rate.Because the prediction result of Random Forest model is chosen in a vote by all decision trees for forming random forest
, and the training sample that while constructing single decision tree uses is obtained by packed method, i.e., has from all samples and put back to
Equiprobability grab sample.If most class number of samples in training set are much larger than minority class sample, very likely due to
Do not include in the training set of single decision tree or wrap containing only a small amount of minority class sample, and leads to the mistake point of minority class sample.Point
Layer Random Forest model thes improvement is that, when constructing the training sample of single decision tree, first with packed method from less
It is sampled in several classes of samples, then there is the equiprobability put back to randomly select and minority class sample size phase from all most class samples
Same sample can not only reduce the mistake point of minority class sample in this way, but also can be to avoid the loss of learning of most class samples.
Preferably, described in step S4 to stratified random forest model optimize especially by construct respectively several by
The stratified random forest model that different decision tree numbers and node number are constituted, with the minimum standard of error outside bag, selection is most
Excellent model.
Preferably, step S5 assesses model prediction result using premeasure curve, the horizontal seat of premeasure curve
It is designated as cumulative area percentage, ordinate is the known mine point number predicted.Curve illustrates mould closer to coordinate system upper left side
Type can use Area Prediction as small as possible mine point as much as possible, have higher forecasting efficiency.
Preferably, step S5 is averaged reduction amount using GINI value as index, carries out importance assessment to each ore information.
Because when metallogenic prognosis, it will usually a variety of ore informations are used, a variety of ore informations are not quite similar at the indicative function of mine, because
And each ore information importance is assessed, the geological information with mineralization correlation maximum is found out, has important meaning to metallogenic prognosis
Justice.
Compared with prior art, the beneficial effects of the present invention are:
1) Copper-nickel Sulfide Ore Deposit metallogenic prognosis is carried out using stratified random forest model, it is possible to reduce mine point mistake minute mentions
High predictablity rate；
2) using geochemical elements paragenetic association information as ore information, it can more accurately reflect different chemistry members
Inner link between element；
3) selection range for determining non-mine point is analyzed using spatial point patterns, it can be ensured that ground locating for non-mine point and mine point
There are notable differences for matter background, increase sample reliability.
Detailed description of the invention
Fig. 1 is flow chart of the invention；
Fig. 2 is the relational graph of stratified random forest model the bag outer error and decision tree number of embodiment building；
Fig. 3 is the research area minerogenetic prognostic map of embodiment；
Fig. 4 is the different random forest model premeasure curve graph of embodiment building；
Fig. 5 is the prospecting factor Assessment of Important figure of embodiment.
Specific embodiment
The attached figures are only used for illustrative purposes and cannot be understood as limitating the patent；
Below in conjunction with drawings and examples, the present invention is further elaborated.
Embodiment 1
As shown in Figure 1, a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model, including such as
Lower step:
Step 1, polynary geologic information in collecting zone, establishes Copper-nickel Sulfide Ore Deposit geographc information data base.
The present embodiment has chosen research ten thousand regional geologic map of area 1:500, ten thousand greenstone zone distribution map of 1:100, ten thousand region 1:100
Geochemistry data and region Copper-nickel Sulfide Ore Deposit distribution map.
Step 2, the Copper-nickel Sulfide Ore Deposit regularity of ore formation in analyzed area extracts ore information.
The present embodiment has chosen 120 Copper-nickel Sulfide Ore Deposits, and using spatial analysis functions to its metallogenetic geologic setting
Comprehensive analysis has been carried out, following ore information is finally extracted: (1) having controlled ore-rock magmatic rock information, control ore-rock magmatic rock in research area is main
It is mafic-ultramafic rocks, has close relationship with the spatial distribution of Copper-nickel Sulfide Ore Deposit；(2) ore-controlling structure information is broken
The channel that construction is often magma and hydrothermal activity is split, equally has apparent coal controlling, therefore be extracted area from database
Main faults in domain, and calculate the distance between each mine point and closest fracture；(3) paragenetic association of geochemistry element is believed
Breath, the more phase property and diversity of mineralizing process, which determine geochemical elements often, has certain inner link, therefore utilizes
Principal component analysis carries out principal component analysis to 15 kinds of elements such as Ni, S, MgO and compound, obtains 9 principal components, it is final choose with
Mineralising it is closely related the 1st, the 2nd common factor is as prospecting factor.
Step 3, non-mine point is chosen, and known mine point is combined to construct training sample set, training stratified random forest model；
Step 3-1 chooses non-mine point.
To guarantee that the non-mine point chosen has different geologic settings from mine point, firstly, known to 120 whole to research area
Mine point carries out spatial point patterns analysis, obtains each mine point with it closest to mine point distance di(i=1,2 ... ... 120)；Its
It is secondary, using each mine point as the center of circle, diMiddle maximum value dmaxFor radius, buffer zone analysis is carried out, gained region is a1, remaining area is
a2；Finally, in region a2In randomly select m point as non-mine point.In the present embodiment, a2The area in region is a1The 4 of area
Times, to guarantee that non-mine point and mine point have identical sampling density, the number for choosing non-mine point m is 4 times, i.e., 480 of n.
Step 3-2 constructs training sample set.
The non-mine point of selection is merged with mine point, generates training sample set, chats point is positive sample, is assigned a value of 1, non-mine
Point is negative sample, is assigned a value of 0.
Step 3-3, training stratified random forest model.
It is concentrated with the positive and negative sample training decision tree that the equiprobability put back to randomly selects identical quantity from sample, repeats this mistake
Cheng Jianli stratified random forest, in the present embodiment, negative sample number is 480, and positive sample number is 120, gloomy using stratified random
Woods model can guarantee respectively to select 120 positive and negative sample training decision trees every time, and sample imbalance is avoided to make model training precision
At influence.In addition, the present embodiment also uses Random Forest model to be trained identical sample, to compare the pre- of different models
Survey result.
Step 4, stratified random forest model is optimized, and carries out metallogenic prognosis using Optimized model；
There are two the factors for influencing stratified random forest precision of prediction: the number of decision tree in stratified random forest；Layering
The number of single decision tree nodes in random forest.
Firstly, the stratified random forest model that building decision tree number is 10 to 5000 respectively, counts different decision trees
The outer error of bag of number stratified random forest models, when error minimum, corresponding decision tree number was optimized parameter, as shown in Figure 2.
Secondly, increase single decision tree nodes number in stratified random forest model one by one, and outside model bag when error minimum, decision burl
Point number is optimal node number, as shown in table 1.In Fig. 2 as can be seen that when decision tree number is greater than 3000, the bag of model
Outer error tends towards stability, thus 3000 be model optimum decision tree number.Equally, table 1 can be seen that the optimizing decision of model
Tree node number is 2.
The relationship of table 1 stratified random forest bag outer error and decision tree nodes number
Finally, carrying out metallogenic prognosis to research area's Copper-nickel Sulfide Ore Deposit using optimization stratified random forest model, obtain
At mine probability graph, and respectively to study the 10% of area's gross area, 50% is boundary, will research zoning be divided into high REGION OF WATER INJECTION OILFIELD, in dive
Power area and low REGION OF WATER INJECTION OILFIELD, as shown in Figure 3.Most of Copper-nickel Sulfide Ore Deposit is all located at into mine height in figure as can be seen that region
REGION OF WATER INJECTION OILFIELD illustrates that prediction result and the goodness of fit of known mine point are very good.In addition, still there is part not find mine at the high REGION OF WATER INJECTION OILFIELD of mine
Point can be used as the key area of next step work of exploration and development.
Step 5, prediction result is verified, ore information importance is assessed.
Firstly, the prediction result according to traditional Random Forest model and stratified random forest respectively, draws different models
Premeasure curve, as shown in Figure 4.As can be seen that the premeasure curve of stratified random forest model is closer to coordinate system in figure
Upper left side illustrates that stratified random forest prediction effect is better than traditional Random Forest model.
Secondly, being averaged reduction amount as index using GINI value, importance assessment is carried out to each ore information, as a result such as Fig. 5 institute
Show.As can be seen that studying each prospecting factor importance ranking in area in figure are as follows: the 1st principal component > 2 principal components > magmatic rock > structure
It makes.
Obviously, the above embodiment of the present invention be only to clearly illustrate example of the present invention, and not be pair
The restriction of embodiments of the present invention.For those of ordinary skill in the art, may be used also on the basis of the above description
To make other variations or changes in different ways.There is no necessity and possibility to exhaust all the enbodiments.It is all this
Made any modifications, equivalent replacements, and improvements etc., should be included in the claims in the present invention within the spirit and principle of invention
Protection scope within.
Claims (8)
1. a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model, which is characterized in that including with
Lower step:
S1: polynary geologic information in collecting zone establishes Copper-nickel Sulfide Ore Deposit geographc information data base；
S2: the Copper-nickel Sulfide Ore Deposit regularity of ore formation in analyzed area extracts ore information；
S3: choosing non-mine point, and known mine point is combined to construct training sample set, training stratified random forest model；
S4: optimizing stratified random forest model, and carries out metallogenic prognosis using Optimized model；
S5: verifying prediction result assesses ore information importance.
2. a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis side based on stratified random forest model according to claim 1
Method, which is characterized in that ore information described in step S2 includes controlling ore-rock magmatic rock information, ore-controlling structure information, geochemistry element to be total to
Raw combined information.
3. a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis side based on stratified random forest model according to claim 2
Method, which is characterized in that the geochemical elements paragenetic association information extracting method is Principal Component Analysis.
4. a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis side based on stratified random forest model according to claim 1
Method, which is characterized in that non-mine point selection range described in step S3 is analyzed by spatial point patterns and determined.
5. a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis side based on stratified random forest model according to claim 1
Method, which is characterized in that training sample set described in step S3 is non-equilibrium sample set, and the non-mine point number studied in area is much larger than mine
Point number.
6. a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis side based on stratified random forest model according to claim 1
Method, which is characterized in that stratified random forest model is optimized especially by constructing several respectively by not described in step S4
It is selected optimal with the stratified random forest model that decision tree number and node number are constituted with the minimum standard of error outside bag
Model.
7. a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis side based on stratified random forest model according to claim 1
Method, which is characterized in that step S5 assesses model prediction result using premeasure curve, and the abscissa of premeasure curve is
Cumulative area percentage, ordinate are the known mine point number predicted.
8. a kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis side based on stratified random forest model according to claim 1
Method, which is characterized in that step S5 is averaged reduction amount using GINI value as index, carries out importance assessment to each ore information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811354629.0A CN109711597A (en) | 2018-11-14 | 2018-11-14 | A kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811354629.0A CN109711597A (en) | 2018-11-14 | 2018-11-14 | A kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109711597A true CN109711597A (en) | 2019-05-03 |
Family
ID=66254787
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811354629.0A Pending CN109711597A (en) | 2018-11-14 | 2018-11-14 | A kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109711597A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110414738A (en) * | 2019-08-01 | 2019-11-05 | 吉林高分遥感应用研究院有限公司 | A kind of crop yield prediction technique and system |
| CN110442666A (en) * | 2019-08-02 | 2019-11-12 | 中国地质调查局发展研究中心 | A kind of mineral resource prediction method and system based on neural network model |
| CN110717618A (en) * | 2019-09-11 | 2020-01-21 | 自然资源部第一海洋研究所 | Submarine hydrothermal sulfide resource evaluation and prediction method based on multi-index comprehensive elements |
| CN110728402A (en) * | 2019-10-10 | 2020-01-24 | 辽宁工程技术大学 | Regional mineral resource prediction system based on geological sample analysis |
| CN111260201A (en) * | 2020-01-13 | 2020-06-09 | 北京科技大学 | Variable importance analysis method based on hierarchical random forest |
| CN111709189A (en) * | 2020-06-22 | 2020-09-25 | 中国地质科学院矿产资源研究所 | Copper ore deposit scale prediction method and system |
| CN111753467A (en) * | 2020-06-23 | 2020-10-09 | 中山大学 | Mineral product result prediction method and device, computer equipment and storage medium |
| CN112489736A (en) * | 2020-12-09 | 2021-03-12 | 中国石油大学(北京) | Mineral content analysis method, device, equipment and storage medium |
| CN113588919A (en) * | 2021-07-23 | 2021-11-02 | 昆明理工大学 | Prediction method suitable for ore prospecting of granite type rare metal deposit |
| CN113656980A (en) * | 2021-08-26 | 2021-11-16 | 中国地质科学院地质力学研究所 | Method and system for determining mining property of fracture area |
| CN114139819A (en) * | 2021-12-07 | 2022-03-04 | 成都理工大学 | Geochemical variable space prediction method based on geostatistical weighted random forest |
| CN115081685A (en) * | 2022-05-30 | 2022-09-20 | 成都理工大学 | Three-dimensional visual positioning prediction method for metal deposit deep resource |
| CN115859106A (en) * | 2022-12-05 | 2023-03-28 | 中国地质大学（北京） | Mineral exploration method and device based on semi-supervised learning and storage medium |
| CN115907151A (en) * | 2022-11-21 | 2023-04-04 | 自然资源陕西省卫星应用技术中心 | Intelligent mineralization prediction method based on geological big data |
| CN116090662A (en) * | 2023-03-02 | 2023-05-09 | 中国地质科学院矿产资源研究所 | Knowledge-graph-based method and system for predicting potential of copper mine outside environment and electronic equipment |
| CN116151107A (en) * | 2023-02-02 | 2023-05-23 | 中国地质大学（北京） | Method, system and electronic equipment for identifying ore potential of magma type nickel cobalt |
| CN117370899A (en) * | 2023-12-08 | 2024-01-09 | 中国地质大学（武汉） | Ore control factor weight determining method based on principal component-decision tree model |
| CN117610733A (en) * | 2023-12-04 | 2024-02-27 | 中国地质大学（北京） | Mineral product prediction method, device, equipment and medium based on deep forest |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706589A (en) * | 2009-12-07 | 2010-05-12 | 中国科学院新疆生态与地理研究所 | Geographic tuple based quantitative prediction method of ore concentration areas |
| RU2013132991A (en) * | 2013-07-16 | 2015-01-27 | Федеральное Государственное Унитарное Предприятие "Сибирский Научно-Исследовательский Институт Геологии, Геофизики И Минерального Сырья" | METHOD FOR FORECASTING HIDDEN ORE MINING ASSOCIATED WITH GRANITOIDS |
-
2018
- 2018-11-14 CN CN201811354629.0A patent/CN109711597A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706589A (en) * | 2009-12-07 | 2010-05-12 | 中国科学院新疆生态与地理研究所 | Geographic tuple based quantitative prediction method of ore concentration areas |
| RU2013132991A (en) * | 2013-07-16 | 2015-01-27 | Федеральное Государственное Унитарное Предприятие "Сибирский Научно-Исследовательский Институт Геологии, Геофизики И Минерального Сырья" | METHOD FOR FORECASTING HIDDEN ORE MINING ASSOCIATED WITH GRANITOIDS |
Non-Patent Citations (3)
| Title |
|---|
| 季斌: "内蒙古浩布高地区多金属矿综合信息找矿预测研究", 《中国优秀博士学位论文全文数据库 基础科学辑》 * |
| 秦昆: "《GIS空间分析理论与方法》", 31 December 2010 * |
| 马海荣 等: "一种处理非平衡数据集的优化随机森林分类方法", 《微电子学与计算机》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110414738B (en) * | 2019-08-01 | 2022-03-22 | 吉林高分遥感应用研究院有限公司 | Crop yield prediction method and system |
| CN110414738A (en) * | 2019-08-01 | 2019-11-05 | 吉林高分遥感应用研究院有限公司 | A kind of crop yield prediction technique and system |
| CN110442666A (en) * | 2019-08-02 | 2019-11-12 | 中国地质调查局发展研究中心 | A kind of mineral resource prediction method and system based on neural network model |
| CN110442666B (en) * | 2019-08-02 | 2021-08-24 | 中国地质调查局发展研究中心 | Mineral resource prediction method and system based on neural network model |
| CN110717618A (en) * | 2019-09-11 | 2020-01-21 | 自然资源部第一海洋研究所 | Submarine hydrothermal sulfide resource evaluation and prediction method based on multi-index comprehensive elements |
| CN110728402A (en) * | 2019-10-10 | 2020-01-24 | 辽宁工程技术大学 | Regional mineral resource prediction system based on geological sample analysis |
| CN110728402B (en) * | 2019-10-10 | 2023-11-03 | 辽宁工程技术大学 | Regional mineral resource prediction system based on geological specimen analysis |
| CN111260201A (en) * | 2020-01-13 | 2020-06-09 | 北京科技大学 | Variable importance analysis method based on hierarchical random forest |
| CN111260201B (en) * | 2020-01-13 | 2023-04-28 | 北京科技大学 | Variable importance analysis method based on layered random forest |
| CN111709189A (en) * | 2020-06-22 | 2020-09-25 | 中国地质科学院矿产资源研究所 | Copper ore deposit scale prediction method and system |
| CN111753467A (en) * | 2020-06-23 | 2020-10-09 | 中山大学 | Mineral product result prediction method and device, computer equipment and storage medium |
| CN111753467B (en) * | 2020-06-23 | 2023-08-22 | 中山大学 | Mineral product result prediction method, device, computer equipment and storage medium |
| CN112489736A (en) * | 2020-12-09 | 2021-03-12 | 中国石油大学(北京) | Mineral content analysis method, device, equipment and storage medium |
| CN113588919A (en) * | 2021-07-23 | 2021-11-02 | 昆明理工大学 | Prediction method suitable for ore prospecting of granite type rare metal deposit |
| CN113656980A (en) * | 2021-08-26 | 2021-11-16 | 中国地质科学院地质力学研究所 | Method and system for determining mining property of fracture area |
| CN114139819A (en) * | 2021-12-07 | 2022-03-04 | 成都理工大学 | Geochemical variable space prediction method based on geostatistical weighted random forest |
| CN114139819B (en) * | 2021-12-07 | 2023-05-23 | 成都理工大学 | Geochemical variable space prediction method based on geostatistically weighted random forest |
| CN115081685A (en) * | 2022-05-30 | 2022-09-20 | 成都理工大学 | Three-dimensional visual positioning prediction method for metal deposit deep resource |
| CN115907151A (en) * | 2022-11-21 | 2023-04-04 | 自然资源陕西省卫星应用技术中心 | Intelligent mineralization prediction method based on geological big data |
| CN115859106A (en) * | 2022-12-05 | 2023-03-28 | 中国地质大学（北京） | Mineral exploration method and device based on semi-supervised learning and storage medium |
| CN116151107A (en) * | 2023-02-02 | 2023-05-23 | 中国地质大学（北京） | Method, system and electronic equipment for identifying ore potential of magma type nickel cobalt |
| CN116151107B (en) * | 2023-02-02 | 2023-09-05 | 中国地质大学（北京） | Method, system and electronic equipment for identifying ore potential of magma type nickel cobalt |
| CN116090662A (en) * | 2023-03-02 | 2023-05-09 | 中国地质科学院矿产资源研究所 | Knowledge-graph-based method and system for predicting potential of copper mine outside environment and electronic equipment |
| CN117610733A (en) * | 2023-12-04 | 2024-02-27 | 中国地质大学（北京） | Mineral product prediction method, device, equipment and medium based on deep forest |
| CN117370899A (en) * | 2023-12-08 | 2024-01-09 | 中国地质大学（武汉） | Ore control factor weight determining method based on principal component-decision tree model |
| CN117370899B (en) * | 2023-12-08 | 2024-02-20 | 中国地质大学（武汉） | Ore control factor weight determining method based on principal component-decision tree model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109711597A (en) | A kind of Copper-nickel Sulfide Ore Deposit metallogenic prognosis method based on stratified random forest model | |
| CN109102180B (en) | Comprehensive parameter evaluation method for double-dessert evaluation of tight sandstone reservoir | |
| CN107807407B (en) | A kind of petroleum zone efficiency evaluation method and apparatus | |
| CN100465998C (en) | Method for finding mine with cubic forecast model based on three-dimensional modeling | |
| CN104615862B (en) | The method that high water cut oil field based on evolution algorithm determines well location | |
| CN106501854B (en) | Oil gas drilling target geological quantifying risk evaluation method | |
| CN103993871B (en) | Method and device for processing well logging information of thin interbed stratums in standardization mode | |
| CN109388817A (en) | A kind of Reservoir Fracture three-dimensional modeling method | |
| CN104318109A (en) | Shale gas reservoir identification method based on support vector machine | |
| CN106355011A (en) | Geochemical data element sequence structure analysis method and device | |
| CN108333629A (en) | A method of it is thick using empirical mode decomposition and support vector machines quantitative forecast coal | |
| CN108663714A (en) | A kind of sedimentary micro depicting method | |
| CN110412662A (en) | Method for prediction of reservoirs of thin interbeded based on seismic multi-attribute deep learning | |
| CN105572747A (en) | Method for identifying lithology of high-salt potassium-rich region water-flooding oil reservoir clastic rocks | |
| CN103628816B (en) | The method of seismic properties Guidance Levels well guiding and device thereof | |
| CN109614741A (en) | The strongly regional sandstone-type uranium deposit Predication of Prospective Areas method of neotectonics activity | |
| Martin et al. | PRISE validates resource triangle concept | |
| Liu et al. | NMR log response prediction from conventional petrophysical logs with XGBoost-PSO framework | |
| CN106199754B (en) | Oil gas drilling target integrates optimizing evaluation method | |
| CN116047590A (en) | Method and device for building earthquake phase prediction model and method and device for predicting earthquake phase | |
| CN108805158A (en) | A kind of fine and close oily reservoir diagenetic phase division methods | |
| Hood et al. | Assessing the Mesaverde basin-centered gas play, Piceance Basin, Colorado | |
| CN103048688A (en) | Seismic attribute optimal selection method based on three-step method | |
| CN111638550A (en) | Fracture-controlled-storage evaluation method | |
| CN105549115A (en) | Method for utilizing logging, well logging and oil test data to quantitatively describe oil-gas distribution |
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: 20190503 |