☆ 4.7 Article

Web-based machine learning tool that determines the origin of natural gases

COMPUTERS & GEOSCIENCES (2020)

Journal

COMPUTERS & GEOSCIENCES

Volume 145, Issue -, Pages -

Publisher

PERGAMON-ELSEVIER SCIENCE LTD

DOI: 10.1016/j.cageo.2020.104595

Keywords

Natural gas; Gas origin; Methane; Machine learning; Classification; Random forest

Categories

Computer Science, Interdisciplinary Applications Geosciences, Multidisciplinary

Ask authors/readers for more resources

Protocol

Community support

Reagent

Community support

Abstract

Investigations on the origin of natural gases traditionally involve manual plotting of values for various geochemical parameters on binary gas genetic diagrams and the comparison of these values with empirically defined gas genetic fields. However, these fields considerably overlap, and the accuracy and uncertainty on the derived origin are not quantified. To overcome these issues, we developed a web-based tool powered by a machine learning model that determines the origin of natural gases. The utilized large global dataset of natural gases (27,852 samples) includes 10,937 samples which we manually interpreted and labeled with one of the five gas origins (thermogenic, primary microbial from CO2 reduction, primary microbial from methyl-type fermentation, secondary microbial, and abiotic). The supervised machine learning model uses random forest algorithm to classify natural gas samples based on four features (geochemical parameters CH4/(C2H6+C3H8), delta C-13-CH4, delta H-2-CH4 and (delta C-13-CO2). The model determines the origin of gases in samples with unknown origin accompanied by model accuracy and the confidence score for each possible origin. The model is deployed on the website www.gasorigin.com with a simple user-friendly interface. The incorporation of more data, geochemical parameters (model features) and determination of post-generation processes are the subjects of future developments.

Authors

I am an author on this paper

Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7

Not enough ratings

Secondary Ratings

Novelty

-

Significance

-

Scientific rigor

-

Rate this paper

Recommended

Article Green & Sustainable Science & Technology

Chemical signatures to identify the origin of solid ashes for efficient recycling using machine learning

Chongchong Qi, Mengting Wu, Xinhang Xu, Qiusong Chen

Summary: Solid ashes are a major environmental problem worldwide, and their efficient recycling requires accurate origin information. In this study, a machine learning-based ash origin detection system was developed using a diverse dataset of 310 solid ash samples from different origins. The system achieved high accuracy in predicting the origins of solid ashes, even without expert knowledge of the operating conditions.

JOURNAL OF CLEANER PRODUCTION (2022)

Add to Collection

Article Environmental Sciences

Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal Wetlands

Ricardo Martinez Prentice, Miguel Villoslada Pecina, Raymond D. Ward, Thaisa F. Bergamo, Chris B. Joyce, Kalev Sepp

Summary: In the study, high-resolution images obtained by multispectral cameras mounted on Unmanned Aerial Vehicles were used to classify coastal wetland sites. The Random Forest classifier outperformed the K-Nearest Neighbors algorithm, especially in pixel-based classification. The findings suggest that for heterogeneous environments like wetlands, pixel-based classification provides a more realistic interpretation of plant community distribution.

REMOTE SENSING (2021)

Add to Collection

Article Environmental Sciences

Acoustic Classification of Juvenile Pacific Salmon (Oncorhynchus spp) and Pacific Herring (Clupea pallasii) Schools Using Random Forests

Shani Rousseau, Stephane Gauthier, Chrys Neville, Stewart Johnson, Marc Trudel

Summary: Acoustic surveys are commonly used to evaluate fish stocks worldwide. This study successfully distinguished juvenile Pacific salmon from Pacific herring aggregations using a random forest approach based on acoustic and morphological characteristics. The model achieved a 98% accuracy in differentiating the two species, with school depth and school mean volume backscattering strength as the most important predictors.

FRONTIERS IN MARINE SCIENCE (2022)

Add to Collection

Article Environmental Sciences

Classifying Forest Type in the National Forest Inventory Context with Airborne Hyperspectral and Lidar Data

Caileigh Shoot, Hans-Erik Andersen, L. Monika Moskal, Chad Babcock, Bruce D. Cook, Douglas C. Morton

Summary: Forest structure and composition play a vital role in regulating various ecosystem services, and forest type information is crucial for quantifying resources and supporting ecological analysis and management decisions. This study developed a methodology using airborne hyperspectral and lidar data to map FIA-defined forest types in interior Alaska, with the random forest classification algorithm showing the highest accuracy. It highlights the benefits of combining structural (lidar) and spectral (imagery) data for forest type classification.

REMOTE SENSING (2021)

Add to Collection

Article Materials Science, Multidisciplinary

Studies on the regularity of perovskite formation via machine learning

Long Li, Qiuling Tao, Pengcheng Xu, Xue Yang, Wencong Lu, MinJie Li

Summary: In this study, a random forest classification (RFC) model was constructed to accurately determine the formability of ABX3 and A2B'B''X6 compounds in perovskite structures. The model achieved high accuracy rates of 96.55% for ABX3 and 91.83% for A2B'B''X6 compounds. Additionally, the model successfully identified 241 ABX3 and 1131 A2B'B''X6 perovskites with high probabilities of formability from a large pool of candidate compounds.

COMPUTATIONAL MATERIALS SCIENCE (2021)

Add to Collection

Article Agriculture, Dairy & Animal Science

Applying Machine Learning Algorithms for the Classification of Mink Infected with Aleutian Disease Using Different Data Sources

Duy Ngoc Do, Guoyu Hu, Pourya Davoudi, Alimohammad Shirzadifar, Ghader Manafiazar, Younes Miar

Summary: Aleutian disease (AD) is a common infectious disease in mink farms, causing financial losses to the mink industry. Predicting AD infected mink without using expensive methods like counterimmunoelectrophoresis (CIEP) can be important for controlling AD. This study demonstrates that the random forest algorithm can accurately classify AD-infected mink, providing a potential solution for implementing machine learning in AD control.

ANIMALS (2022)

Add to Collection

Article Materials Science, Multidisciplinary

Machine Learning Methods for the Prediction of the Inclusion Content of Clean Steel Fabricated by Electric Arc Furnace and Rolling

Estela Ruiz, Diego Ferreno, Miguel Cuartas, Lara Lloret, Pablo M. Ruiz del arbol, Ana Lopez, Francesc Esteve, Federico Gutierrez-Solana

Summary: Machine learning classification models were used to analyze experimental data of clean cold forming steel, identifying variables influencing inclusion content and interpreting their impact through Permutation Importance and Partial Dependence Plots. The study found that changes induced by hot rolling in coil diameter and other metallurgical operations significantly affect inclusion cleanliness, and recommendations were made to optimize sampling processes and manufacturing conditions to improve steel inclusion cleanliness.

METALS (2021)

Add to Collection

Article Environmental Sciences

Comparing Machine Learning Algorithms for Pixel/Object-Based Classifications of Semi-Arid Grassland in Northern China Using Multisource Medium Resolution Imageries

Nitu Wu, Luis Guilherme Teixeira Crusiol, Guixiang Liu, Deji Wuyun, Guodong Han

Summary: Timely and accurate grassland classification is crucial for grassland resource management. However, there is a lack of comparative studies on commonly used methods for semi-arid grasslands in northern China. This study compared the performance of four machine learning algorithms for mapping semi-arid grassland using pixel-based and object-based classification methods. The findings showed that the object-based methods provided a more realistic land cover distribution and higher accuracy.

REMOTE SENSING (2023)

Add to Collection

Article Food Science & Technology

1H NMR-based metabolomic approach combined with machine learning algorithm to distinguish the geographic origin of huajiao (Zanthoxylum bungeanum Maxim.)

Chuanjian Cui, Mingyue Xia, Ziqi Wei, Jianglin Chen, Chuanyi Peng, Huimei Cai, Long Jin, Ruyan Hou

Summary: In this study, 1H NMR spectroscopy was used for the first time to analyze the geographic origin of Chinese prickly ash. A classification model based on machine learning algorithms achieved high accuracy in distinguishing the origin of the spice. The main marker compounds responsible for differentiation were identified as linalool, linalyl acetate, nonanal, and ocimene.

FOOD CONTROL (2023)

Add to Collection

Article Environmental Sciences

Superpixel segmentation integrated feature subset selection for wetland classification over Yellow River Delta

Long Cui, Jiahua Zhang, Zhenjiang Wu, Lan Xun, Xiaopeng Wang, Shichao Zhang, Yun Bai, Sha Zhang, Shanshan Yang, Qi Liu

Summary: Wetlands in the Yellow River Delta are important and vulnerable due to tidal action and sediment deposits. A object-oriented approach with feature preference machine learning was used to classify the wetlands. A superpixel segmentation method using the watershed algorithm improved the classification accuracy. The random forest classifier combining superpixel segmentation and feature selection methods outperformed other pixel-based machine learning methods with a 91.74% overall accuracy and a kappa coefficient of 0.9078.

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH (2023)

Add to Collection

Article Environmental Sciences

A Spatial Data-Driven Approach for Mineral Prospectivity Mapping

Indishe P. Senanayake, Anthony S. Kiem, Gregory R. Hancock, Vaclav Metelka, Chris B. Folkes, Phillip L. Blevin, Anthony R. Budd

Summary: Mineral prospectivity mapping is crucial for discovering new economic mineral deposits, but it often requires significant costs, time, and human resources. This study utilized an ensemble machine learning approach with geoscience datasets to map Cu-Au and Pb-Zn mineral prospectivity in the Cobar Basin, NSW, Australia. The results showed improved classification accuracy compared to existing methods and demonstrated the potential for this approach to serve as a preliminary evaluation technique, providing guidance for more detailed geological investigations in other regions.

REMOTE SENSING (2023)

Add to Collection

Article Genetics & Heredity

RF_phage virion: Classification of phage virion proteins with a random forest model

Yanqin Zhang, Zhiyuan Li

Summary: In this study, machine learning methods were used to classify phage virion proteins, and a novel approach called RF_phage virion was proposed for effective classification. The model utilized four protein sequence coding methods as features and employed the random forest algorithm for classification. The results showed that the RF_phage virion model outperformed classical machine learning methods.

FRONTIERS IN GENETICS (2023)

Add to Collection

Article Computer Science, Artificial Intelligence

Identifying patients with paroxysmal atrial fibrillation from sinus rhythm ECG using random forests

Evangelia Myrovali, Dimitrios Hristu-Varsakelis, Dimitrios Tachmatzidis, Antonios Antoniadis, Vassilios Vassilikos

Summary: This paper presents a method using sinus rhythm electrocardiogram (ECG) recordings to identify patients with a history of paroxysmal atrial fibrillation (PAF). By analyzing key ECG metrics of P-waves, including novel amplitude and slope-based features, the method achieved high accuracy, sensitivity, and specificity for classifying PAF patients. The approach has potential value for the early identification of individuals prone to PAF episodes.

EXPERT SYSTEMS WITH APPLICATIONS (2023)

Add to Collection

Article Energy & Fuels

A data-driven shale gas production forecasting method based on the multi-objective random forest regression

Liang Xue, Yuetian Liu, Yifei Xiong, Yanli Liu, Xuehui Cui, Gang Lei

Summary: The study proposes a multi-objective random forest method to predict dynamic shale gas production data, utilizing geological and hydraulic fracturing properties as predictive input features. By determining appropriate hyperparameters, considering variable importance, and incorporating features such as initial peak production rate, the predictive accuracy can be significantly improved. Additionally, it is found that using more sample data with less measurement errors can increase the accuracy of the data-driven shale gas production model.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2021)

Add to Collection

Article Biochemical Research Methods

SNARER: new molecular descriptors for SNARE proteins classification

Alessia Auriemma Citarella, Luigi Di Biasi, Michele Risi, Genoveffa Tortora

Summary: This study introduces a new class of molecular descriptors (SNARER) related to the chemical-physical properties of proteins to enhance the classification performance of SNARE proteins. Results show that the addition of SNARER descriptors improves precision for all machine learning algorithms tested. Additionally, it is highlighted that using a balanced dataset can significantly enhance forecast accuracy.

BMC BIOINFORMATICS (2022)

Add to Collection

Article Geosciences, Multidisciplinary

Formation and preservation of a giant petroleum accumulation in superdeep carbonate reservoirs in the southern Halahatang oil field area, Tarim Basin, China

Guangyou Zhu, Alexei V. Milkov, Zhiyao Zhang, Chonghao Sun, Xiaoxiao Zhou, Feiran Chen, Jianfa Han, Yongfeng Zhu

AAPG BULLETIN (2019)

Add to Collection

Article Energy & Fuels

Diamondoids as tracers of late gas charge in oil reservoirs: Example from the Tazhong area, Tarim Basin, China

Guangyou Zhu, Zhiyao Zhang, Alexei V. Milkov, Xiaoxiao Zhou, Haijun Yang, Jianfa Han

FUEL (2019)

Add to Collection

Article Geosciences, Multidisciplinary

Randomness, serendipity, and luck in petroleum exploration

Alexei Milkov, William C. Navidi

AAPG BULLETIN (2020)

Add to Collection

Article Geosciences, Multidisciplinary

Turning dry holes from disasters to exploration wisdom: Decision tree to determine the key failure mode for segments in conventional petroleum prospects

Alexei Milkov, Jack M. Samis

AAPG BULLETIN (2020)

Add to Collection

Article Geosciences, Multidisciplinary

Thickness matters: Influence of dolerite sills on the thermal maturity of surrounding rocks in a coal bed methane play in Botswana

Muhammed Emin Bulguroglu, Alexei V. Milkov

MARINE AND PETROLEUM GEOLOGY (2020)

Add to Collection

Article Geochemistry & Geophysics

Geochemistry of shale gases from around the world: Composition, origins, isotope reversals and rollovers, and implications for the exploration of shale plays

Alexei Milkov, Mohinudeen Faiz, Giuseppe Etiope

ORGANIC GEOCHEMISTRY (2020)

Add to Collection

Article Multidisciplinary Sciences

Using global isotopic data to constrain the role of shale gas production in recent increases in atmospheric methane

Alexei Milkov, Stefan Schwietzke, Grant Allen, Owen A. Sherwood, Giuseppe Etiope

SCIENTIFIC REPORTS (2020)

Add to Collection

Article Energy & Fuels

Comprehensive Molecular Compositions and Origins of DB301 Crude Oil from Deep Strata, Tarim Basin, China

Meng Wang, Guangyou Zhu, Alexei V. Milkov, Linxian Chi

ENERGY & FUELS (2020)

Add to Collection

Editorial Material Geosciences, Multidisciplinary

Reply to the comment by Iyer and Schmid on Thickness matters: Influence of dolerite sills on the thermal maturity of surrounding rocks in a coal bed methane play in Botswana by Bulguroglu, ME, Milkov, AV, Marine and Petroleum Geology 111 (2020), 219-229 Discussion

Muhammed Emin Bulguroglu, Alexei V. Milkov

MARINE AND PETROLEUM GEOLOGY (2020)

Add to Collection

Article Geosciences, Multidisciplinary

Breathing new life into postmortem analysis: Identification of key failure modes for conventional petroleum segments offshore New Zealand

Jack M. Samis, Alexei Milkov

MARINE AND PETROLEUM GEOLOGY (2020)

Add to Collection

Article Geosciences, Multidisciplinary

FORECASTING ABILITIES OF INDIVIDUAL PETROLEUM EXPLORERS: PRELIMINARY FINDINGS FROM CROWDSOURCED PROSPECT ASSESSMENTS

A. V. Milkov

JOURNAL OF PETROLEUM GEOLOGY (2020)

Add to Collection

Article Energy & Fuels

Deepest oil in Asia: Characteristics of petroleum system in the Tarim basin, China

Guangyou Zhu, Alexei Milkov, Jingfei Li, Nan Xue, Yongquan Chen, Jianfeng Hu, Tingting Li, Zhiyao Zhang, Zhiyong Chen

Summary: The LT1 well in the Tarim Basin of China has reached a record depth in Asia and has yielded significant amounts of oil and gas from the Cambrian Wusongger Formation. With a high pressure of 90.8 Mpa and temperature of 162 degrees C, this discovery provides valuable insights for global exploration of Paleozoic ultra-deep oil.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2021)

Add to Collection

Article Energy & Fuels

Reporting the expected exploration outcome: When, why and how the probability of geological success and success-case volumes for the well differ from those for the prospect

Alexei Milkov

Summary: It is essential for petroleum explorers and exploration companies to differentiate between geological success probability and success-case volumes, accurately report expected outcomes of exploration wells, and understand key differences in geological PoS and success-case volumes. Proper reporting and modeling of drilling outcomes will lead to better decisions and maximize the expected value of exploration projects. Postmortem evaluations should compare well outcomes with pre-drill assessments to ensure accuracy and improve future decision-making.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2021)

Add to Collection

Article Geochemistry & Geophysics

New approaches to distinguish shale-sourced and coal-sourced gases in petroleum systems

V. Alexei Milkov

Summary: Current methods correlating thermogenic gases with shale and coal source rocks are unreliable and not globally applicable. A new approach based on stable carbon isotopes provides a more accurate separation of shale-sourced and coal-sourced gases, which has been tested and validated in several petroleum systems.

ORGANIC GEOCHEMISTRY (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

An advanced median filter for improving the signal-to-noise ratio of seismological datasets

Yapo Abole Serge Innocent Oboue, Yunfeng Chen, Sergey Fomel, Wei Zhong, Yangkang Chen

Summary: Strong noise can disrupt the recorded seismic waves and negatively impact subsequent seismological processes. To improve the signal-to-noise ratio (S/N) of seismological data, we introduce MATamf, an open-source MATLAB code package based on an advanced median filter (AMF) that simultaneously attenuates various types of noise and improves S/N. Experimental results demonstrate the usefulness and advantages of the proposed AMF workflow in enhancing the S/N of a wide range of seismological applications.

COMPUTERS & GEOSCIENCES (2024)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Advection-based tracking and analysis of salinity movement in the Indian Ocean

Upkar Singh, P. N. Vinayachandran, Vijay Natarajan

Summary: The Bay of Bengal maintains its salinity distribution due to the cyclic flow of high salinity water and the mixing with freshwater. This paper introduces an advection-based feature definition and algorithms to track the movement of high salinity water, validated through comparison with observed data.

COMPUTERS & GEOSCIENCES (2024)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Automated mapping of bedrock-fracture traces from UAV-acquired images using U-Net convolutional neural networks

Bijal Chudasama, Nikolas Ovaskainen, Jonne Tamminen, Nicklas Nordback, Jon Engstro, Ismo Aaltonen

Summary: This contribution presents a novel U-Net convolutional neural network (CNN)-based workflow for automated mapping of bedrock fracture traces from aerial photographs acquired by unmanned aerial vehicles (UAV). The workflow includes training a U-Net CNN using a small subset of photographs with manually traced fractures, semantic segmentation of input images, pixel-wise identification of fracture traces, ridge detection algorithm and vectorization. The results show the effectiveness and accuracy of the workflow in automated mapping of bedrock fracture traces.

COMPUTERS & GEOSCIENCES (2024)

Add to Collection

Article Computer Science, Interdisciplinary Applications

A novel finer soil strength mapping framework based on machine learning and remote sensing images

Ruizhen Wang, Siyang Wan, Weitao Chen, Xuwen Qin, Guo Zhang, Lizhe Wang

Summary: This paper proposes a novel framework to generate a finer soil strength map based on RCI, which uses ensemble learning models to obtain USCS soil classification and predict soil moisture, in order to improve the resolution and reliability of existing soil strength maps.

COMPUTERS & GEOSCIENCES (2024)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Intelligent terrain generation considering global information and terrain patterns

Zhanlong Chen, Xiaochuan Ma, Houpu Li, Xuwei Xu, Xiaoyi Han

Summary: Simulated terrains are important for landform and terrain research, disaster prediction, rescue and disaster relief, and national security. This study proposes a deep learning method, IGPN, that integrates global information and pattern features of the local terrain to generate accurate simulated terrains quickly.

COMPUTERS & GEOSCIENCES (2024)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Physics-Informed Neural Networks for solving transient unconfined groundwater flow

Daniele Secci, Vanessa A. Godoy, J. Jaime Gomez-Hernandez

Summary: Neural networks excel in various machine learning applications, but lack physical interpretability and constraints, limiting their accuracy and reliability in predicting complex physical systems' behavior. Physics-Informed Neural Networks (PINNs) integrate neural networks with physical laws, providing an effective tool for solving physical problems. This article explores recent developments in PINNs, emphasizing their application in solving unconfined groundwater flow, and discusses challenges and opportunities in this field.

COMPUTERS & GEOSCIENCES (2024)

Add to Collection

Article Computer Science, Interdisciplinary Applications

A physically constrained hybrid deep learning model to mine a geochemical data cube in support of mineral exploration

Renguang Zuo, Ying Xu

Summary: This study proposes a hybrid deep learning model consisting of a one-dimensional convolutional neural network (1DCNN) and a graph convolutional network (GCN) to extract joint spectrum-spatial features from geochemical survey data for mineral exploration. The physically constrained hybrid model performs better in geochemical anomaly recognition compared to other models.

COMPUTERS & GEOSCIENCES (2024)

Add to Collection

© Peeref 2019-2024. All rights reserved.