With an increasing demand for raw materials, predictive models that support successful mineral exploration targeting are of great importance. We evaluated different machine learning techniques with an emphasis on boosting algorithms and implemented them in an ArcGIS toolbox. Performance was tested on an exploration dataset from the Iberian Pyrite Belt (IPB) with respect to accuracy, performance, stability, and robustness. Boosting algorithms are ensemble methods used in supervised learning for regression and classification. They combine weak classifiers, i.e., classifiers that perform slightly better than random guessing to obtain robust classifiers. Each time a weak learner is added; the learning set is reweighted to give more importance to misclassified samples. Our test area, the IPB, is one of the oldest mining districts in the world and hosts giant volcanic-hosted massive sulfide (VMS) deposits. The spatial density of ore deposits, as well as the size and tonnage, makes the area unique, and due to the high data availability and number of known deposits, well-suited for testing machine learning algorithms. We combined several geophysical datasets, as well as layers derived from geological maps as predictors of the presence or absence of VMS deposits. Boosting algorithms such as BrownBoost and Adaboost were tested and compared to Logistic Regression (LR), Random Forests (RF) and Support Vector machines (SVM) in several experiments. We found performance results relatively similar, especially to BrownBoost, which slightly outperformed LR and SVM with respective accuracies of 0.96 compared to 0.89 and 0.93. Data augmentation by perturbing deposit location led to a 7% improvement in results. Variations in the split ratio of training and test data led to a reduction in the accuracy of the prediction result with relative stability occurring at a critical point at around 26 training samples out of 130 total samples. When lower numbers of training data were introduced accuracy dropped significantly. In comparison with other machine learning methods, Adaboost is user-friendly due to relatively short training and prediction times, the low likelihood of overfitting and the reduced number of hyperparameters for optimization. Boosting algorithms gave high predictive accuracies, making them a potential data-driven alternative for regional scale and/or brownfields mineral exploration.
相似文献 |
|||||||
|
|||||||
|
排序方式： 共有15条查询结果，搜索用时 203 毫秒
1.
Effects of different types of mechanical disturbances on a charophyte dominated macrophyte community
Charophytes are unique and endangered macroalgae. Mechanical stress is known to damage charophyte communities, however, virtually nothing is known how such disturbance affect seasonal succession and recovery of charophyte communities. We examined the response of a shallow water charophyte community to different types and timing of disturbances by conducting seasonal in situ field experiment in a brackish water ecosystem. The disturbances associated with mechanical removal of vegetation or sediment had the largest impact on the charophyte community. Disturbances that only partly damaged the plants had minor effects and communities were characterized by fast recovery. The timing of disturbance had significant effects on the charophyte community with stronger effects when disturbed at the early stage compared to mid-stage of seasonal succession. 相似文献
2.
We investigate the general properties of supernova driven interstellar turbulence using local three-dimensional MHD simulations under Galactic conditions. Our model includes the effects of large-scale shear due to Galactic differential rotation, density stratification, compressibility, magnetic fields, heating via supernova explosions and parameterized radiative cooling of the interstellar medium. In addition to investigating isolated supernova explosions we allow for multiple supernovae distributed randomly in the Galactic disc and exponentially in the vertical direction. Single supernova explosions drive a strong shock, the lifetime of which is approximately 2 Myr in our model. This stage is found to be characterized by a kinetic energy spectrum in the diffuse gas with spectral index consistent with k = –2. Large-scale shear and Coriolis force act on the supernova remnant producing some vorticity inside it, but this process was found to be very weak. In the case of multiple supernova explosions, older remnants have an important role causing density fluctuations in the interstellar medium. In this clumpy medium, the propagation velocity of the shock fronts changes due to the changing density, and vorticity is generated. In the absence of these supernova interactions the kinetic energy spectrum shows a relatively wide shock spectrum with spectral index k = –2, but when the supernova interactions become dominant the classical k = –5/3 spectrum is observed. 相似文献
3.
Brandmeier Melanie Cabrera Zamora Irving Gibran Nykänen Vesa Middleton Maarit 《Natural Resources Research》2020,29(1):71-88
4.
Maarit Middleton Paavo Nrhi Raimo Sutinen 《ISPRS Journal of Photogrammetry and Remote Sensing》2011,66(3):287-297
In a humid northern boreal climate, the success rate of artificial regeneration to Scots pine (Pinus sylvestris L.) can be improved by including a soil water content (SWC) based assessment of site suitability in the reforestation planning process. This paper introduces an application of airborne visible-near-infrared imaging spectroscopic data to identify suitable subregions of forest compartments for the low SWC-tolerant Scots pine. The spatial patterns of understorey plant species communities, recorded by the AISA (Airborne Imaging Spectrometer for Applications) sensor, were demonstrated to be dependant on the underlying SWC. According to the nonmetric multidimensional scaling and correlation results twelve understorey species were found to be most abundant on sites with high soil SWCs. The abundance of bare soil, rocks and abundance of more than ten species indicated low soil SWCs. The spatial patterns of understorey are attributed to time-stability of the underlying SWC patterns. A supervised artificial neural network (radial basis functional link network, probabilistic neural network) approach was taken to classify AISA imaging spectrometer data with dielectric (as a measure volumetric SWC) ground referencing into regimes suitable and unsuitable for Scots pine. The accuracy assessment with receiver operating characteristics curves demonstrated a maximum of 74.1% area under the curve values which indicated moderate success of the NN modelling. The results signified the importance of the training set’s quality, adequate quantity (>2.43 points/ha) and NN algorithm selection over the NN algorithm training parameter optimization to perfection. This methodology for the analysis of site suitability of Scots pine can be recommended, especially when artificial regeneration of former mixed wood Norway spruce (Picea abies L. Karst) - downy birch (Betula pubenscens Ehrh.) stands is being considered, so that artificially regenerated areas to Scots pine can be optimized for forestry purposes. 相似文献
5.
6.
Mira Markovaara-Koivisto Antti E.K. Ojala Jussi Mattila Ilmo Kukkonen Ilkka Aro Arto Pullinen Pekka Hänninen Maarit Middleton Aleksi Sutinen Juha Majaniemi Timo Ruskeeniemi Raimo Sutinen 《地球表面变化过程与地形》2020,45(12):3011-3024
The cyclic nature of glaciations and related postglacial faulting represents a risk for the deep geological disposal of spent nuclear fuel in areas likely to be affected by future glaciations. Seismic history was therefore studied by means of detecting geomorphological structures on airborne laser scanning digital elevation models and underground by excavating in an esker and trenching across a postglacial fault located in northern Fennoscandia. OLS dating and assessing the geomorphological structures was used for timing of the seismic history. The results suggest that the faulting of different segments in the Pasmajärvi complex is due to at least two late Weichselian events, which probably occurred both subglacially and postglacially. The most reliable input for the moment magnitude estimates was vertical slip profiles, and therefore these estimates (MW ≈ 6.4–6.9) are suggested. © 2020 John Wiley & Sons, Ltd. 相似文献
7.
Raimo Sutinen Paavo NÄrhi Maarit Middleton Pekka HÄnninen Mauri Timonen Marja‐Liisa Sutinen 《Boreas: An International Journal of Quaternary Research》2012,41(3):367-378
As a result of global changes, shifts of alpine tree lines towards higher elevations have been recorded, but the role of the spatial variability of the snowpack and zonal‐pattern soil‐nutrient regimes is poorly understood. Norway spruce (Picea abies (L.) Karst) is best suited to fertile soils, and hence we applied soil physical‐chemical and snow measurements and the age chronology of Norway spruce along an elevational gradient (380–557 m a.s.l.) to address a vertical soil zonality hypothesis on mafic Lommoltunturi fell in Finnish Lapland. With regard to increasing elevation, we found an increase in soil NTOT, CTOT and Al, but a decrease in soil Ca, Mg and Ca:Al ratio as well as in electrical conductivity (EC). In addition, the snowpack was significantly thicker in low‐elevation forest than in the tree line and open tundra. In the 1840s, spruce established on low‐elevation soils with a Ca:Al ratio of 2.2. Starting from the 1920s a significant shift of spruce occurred such that it took 60 years to expand the tree line by 55 m in elevation. The spruce tree line has advanced, and the age distribution indicates new colonization of spruce in closed forest up to tundra. The poor soil Ca:Al ratio of 0.02 on tundra apparently is a constraint for spruce. Spruce forest is young (<165 years), and hence we argue that spruce has expanded onto formerly tree‐free sites of this mafic fell. This paper demonstrates that vertical soil zonality is a potential driver for the diffuse tree line of Picea abies on mafic Fennoscandian fells. 相似文献
8.
9.
Maarit J. Korpi 《Astrophysics and Space Science》2004,289(3-4):449-457
Briefly stated, the interstellar matter (hereafter ISM) is a highlyturbulent flow, the turbulence mainly being driven by the violentstirring by supernova explosions, penetrated by dynamicallysignificant magnetic fields both on small and largescales. Sophisticated numerical models of such a turbulent flow cangive important information on some of the complicated processesoccurring in the ISM, e.g. on the regulation of the observedmultiphase structure, properties of interstellar turbulence and thegeneration and maintenance of magnetic fields by the flow itself; inthis paper such models and their most important results are reviewed. 相似文献
10.
We combine a convectively driven dynamo in a spherical shell with a nearly isothermal density-stratified cooling layer that mimics some aspects of a stellar corona to study the emergence and ejections of magnetic field structures. This approach is an extension of earlier models, where forced turbulence simulations were employed to generate magnetic fields. A spherical wedge is used which consists of a convection zone and an extended coronal region to ???1.5 times the radius of the sphere. The wedge contains a quarter of the azimuthal extent of the sphere and 150° in latitude. The magnetic field is self-consistently generated by the turbulent motions due to convection beneath the surface. Magnetic fields are found to emerge at the surface and are ejected to the coronal part of the domain. These ejections occur at irregular intervals and are weaker than in earlier work. We tentatively associate these events with coronal mass ejections on the Sun, even though our model of the solar atmosphere is rather simplistic. 相似文献
|
| 设为首页 | 免责声明 | 关于勤云 | 加入收藏 |
|
Copyright©北京勤云科技发展有限公司 京ICP备09084417号 |