Quantifying the sensitivity of seismic facies classification to seismic attribute selection: An explainable machine-learning study

相 灵敏度(控制系统) 集合(抽象数据类型) 地质学 噪音(视频) 选择(遗传算法) 计算机科学 数据挖掘 任务(项目管理) 机器学习 人工智能 古生物学 工程类 图像(数学) 系统工程 构造盆地 电子工程 程序设计语言
作者
David Lubo-Robles,Deepak Devegowda,Vikram Jayaram,Heather Bedle,Kurt J. Marfurt,Matthew J. Pranter
出处
期刊:Interpretation [Society of Exploration Geophysicists]
卷期号:10 (3): SE41-SE69 被引量:21
标识
DOI:10.1190/int-2021-0173.1
摘要

Abstract During the past two decades, geoscientists have used machine learning (ML) to produce a more quantitative reservoir characterization and discover hidden patterns in their data. However, as the complexity of these models increases, the sensitivity of their results to the choice of the input data becomes more challenging. Measuring how the model uses the input data to perform either a classification or regression task provides an understanding of the data-to-geology relationships which indicates how confident we are in the prediction. To provide such insight, the ML community has developed local interpretable model-agnostic explanations (LIME) and Shapley additive explanations (SHAP) tools. In this study, we train a random forest architecture using a suite of seismic attributes as input to differentiate among mass transport deposits (MTDs), salt, and conformal siliciclastic sediments in a Gulf of Mexico data set. We apply SHAP to understand how the model uses the input seismic attributes to identify target seismic facies and examine in what manner variations in the input, such as adding band-limited random noise or applying a Kuwahara filter, impact the model predictions. During our global analysis, we find that the attribute importance is dynamic, and it changes based on the quality of the seismic attributes and the seismic facies analyzed. For our data volume and target facies, attributes measuring changes in dip and energy show the largest importance for all cases in our sensitivity analysis. We note that to discriminate between the seismic facies, the ML architecture learns a “set of rules” in multiattribute space, and overlap among MTDs, salt, and conformal sediments might exist based on the seismic attribute analyzed. Finally, using SHAP at a voxel scale, we understand why certain areas of interest were misclassified by the algorithm and perform an in-context interpretation to analyze how changes in the geology impacted the model predictions.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
柠檬发布了新的文献求助10
刚刚
凶狠的八宝粥关注了科研通微信公众号
1秒前
孟滴滴的应助被我有我风格采纳,获得50
1秒前
2秒前
2秒前
2秒前
3秒前
4秒前
undo发布了新的文献求助10
4秒前
小Y完成签到,获得积分10
4秒前
郑睿涛发布了新的文献求助10
5秒前
5秒前
ai91完成签到,获得积分10
5秒前
水晶果变萌完成签到 ,获得积分10
5秒前
6秒前
研友_8yNO0L完成签到,获得积分10
6秒前
小章完成签到,获得积分10
6秒前
7秒前
7秒前
7秒前
秋风的应助被corleeang采纳,获得10
7秒前
7秒前
辛夏子完成签到,获得积分10
7秒前
喜悦绮山完成签到,获得积分10
7秒前
7秒前
所所的应助被hahaha625采纳,获得20
8秒前
钟梓袄发布了新的文献求助10
8秒前
yyyyy发布了新的文献求助30
8秒前
scx发布了新的文献求助10
8秒前
9秒前
chen完成签到,获得积分10
9秒前
banlichen完成签到,获得积分10
10秒前
温暖砖头发布了新的文献求助10
10秒前
科目三的应助被n11采纳,获得10
10秒前
Xuan发布了新的文献求助10
10秒前
接受所有饼干完成签到,获得积分10
10秒前
小超完成签到,获得积分10
11秒前
淡定的日记本完成签到,获得积分20
11秒前
12秒前
耍酷谷雪发布了新的文献求助10
12秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
Improving the wear resistance of vehicle gear oil by oil soluble LaF3 nanoparticles and typical additives of gear oil 500
Boundaries and Bodhisattvas 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8183091
求助须知:如何正确求助?哪些是违规求助? 9572406
关于积分的说明 21825003
捐赠科研通 7779812
什么是DOI,文献DOI怎么找? 3433837
关于科研通互助平台的介绍 2558045
邀请新用户注册赠送积分活动 2480046