| 大数据与数学地球科学研究进展——大数据与数学地球科学专题代序 |
| |
| 引用本文： | 周永章,陈铄,张旗,肖凡,王树功,刘艳鹏,焦守涛.大数据与数学地球科学研究进展——大数据与数学地球科学专题代序[J].岩石学报,2018,34(2):255-263. |
| |
| 作者姓名： | 周永章 陈铄 张旗 肖凡 王树功 刘艳鹏 焦守涛 |
| |
| 作者单位： | 广东省地质过程与矿产资源探查重点实验室, 广州 510275;中山大学地球环境与地球资源研究中心, 广州 510275;中山大学地球科学与工程学院, 广州 510275,广东省地质过程与矿产资源探查重点实验室, 广州 510275;中山大学地球环境与地球资源研究中心, 广州 510275;中山大学地球科学与工程学院, 广州 510275,中国科学院地质与地球物理研究所, 北京 100029,广东省地质过程与矿产资源探查重点实验室, 广州 510275;中山大学地球环境与地球资源研究中心, 广州 510275;中山大学地球科学与工程学院, 广州 510275,广东省地质过程与矿产资源探查重点实验室, 广州 510275;中山大学地球环境与地球资源研究中心, 广州 510275;中山大学地球科学与工程学院, 广州 510275,广东省地质过程与矿产资源探查重点实验室, 广州 510275;中山大学地球环境与地球资源研究中心, 广州 510275;中山大学地球科学与工程学院, 广州 510275,广东省地质过程与矿产资源探查重点实验室, 广州 510275;中山大学地球环境与地球资源研究中心, 广州 510275;中山大学地球科学与工程学院, 广州 510275 |
| |
| 基金项目： | 本文受国家重点研发计划项目（2016YFC0600506）、国家自然科学基金项目（41273040）、中国地质调查局项目（12120113067600）和高校基本科研业务费中山大学科研助手资助计划联合资助. |
| |
| 摘 要： | 大数据与数学地球科学的核心应用技术包括高维数据降维、图像数据处理、无限数据流挖掘、机器学习、关联规则算法与推荐系统算法等。人工智能地质学,包括大数据-智能矿床成因模型与找矿模型的构建,是具有重要价值的研究方向。高维数据降维旨在从初始高维特征集合中选出低维特征集合,有效地消除无关和冗余特征,增强学习结果的易理解性。哈希算法、聚类分析、主成分分析等是较常用的数学降维工具。机器学习是人工智能的核心,是使计算机具有智能的根本途径。机器学习与人工智能各种基础问题的统一性观点正在形成。深度学习的训练模型往往需要海量数据作为支撑,因此迁移学习方法日益受到重视。图像模式识别是大数据挖掘的重要技术。网络中的社区结构识别对理解整个网络的结构和功能有重要价值,可帮助分析、预测网络各元素间的交互关系。沉浸式虚拟现实技术是实现大数据可视化的重要方向,对具有多元、异构、时空性、非线性、多尺度地质矿产勘查数据的展示要求有特别的价值。引入VR技术进行矿产地质大数据的可视化,可实现大数据时代矿产勘查数据的新认知。无限数据流在地质、地球化学、地球物理监测中大量存在,甚至可以持续自动产生。对数据流数据的计算包括对点查询、范围查询、内积查询、分位数计算、频繁项计算等。关联规则和推荐系统算法是大数据挖掘中的重要算法,其应用范围越来越广泛。贝叶斯原理在大数据时代有独特的价值,贝叶斯网络是成因建模的一个革命性工具。智能地质学研究刚刚起步,构建大数据-智能矿床成因模型与找矿模型是智能地质学研究的重要内容。矿床模型研究方式的变革,将出现于互联网、云计算技术环境下全球各地的矿床研究团队的共同参与。
|
| 关 键 词： | 大数据挖掘 高维数据降维 图像数据处理 无限数据流挖掘 机器学习 关联规则 人工智能地质学 智能矿床模型 贝叶斯网络 |
| 收稿时间： | 2017/9/10 0:00:00 |
| 修稿时间： | 2017/11/3 0:00:00 |
Advances and prospects of big data and mathematical geoscience |
| |
| ZHOU YongZhang,CHEN Shuo,ZHANG Qi,XIAO Fan,WANG ShuGong,LIU YanPeng and JIAO ShouTao.Advances and prospects of big data and mathematical geoscience[J].Acta Petrologica Sinica,2018,34(2):255-263. |
| |
| Authors: | ZHOU YongZhang CHEN Shuo ZHANG Qi XIAO Fan WANG ShuGong LIU YanPeng and JIAO ShouTao |
| |
| Institution: | Center for Earth Environment & Resources, Sun Yat-senUniversity, Guangzhou 510275, China;Guangdong Provinical Key Laboratory of Mineral Resources and Geological Processes, Guangzhou 510275, China;School of Earth Sciences & Engineering, Sun Yat-senUniversity, Guangzhou 510275, China,Center for Earth Environment & Resources, Sun Yat-senUniversity, Guangzhou 510275, China;Guangdong Provinical Key Laboratory of Mineral Resources and Geological Processes, Guangzhou 510275, China;School of Earth Sciences & Engineering, Sun Yat-senUniversity, Guangzhou 510275, China,Institute of Geology and Geophysics, China Academy of Sciences, Beijing 100029, China,Center for Earth Environment & Resources, Sun Yat-senUniversity, Guangzhou 510275, China;Guangdong Provinical Key Laboratory of Mineral Resources and Geological Processes, Guangzhou 510275, China;School of Earth Sciences & Engineering, Sun Yat-senUniversity, Guangzhou 510275, China,Center for Earth Environment & Resources, Sun Yat-senUniversity, Guangzhou 510275, China;Guangdong Provinical Key Laboratory of Mineral Resources and Geological Processes, Guangzhou 510275, China;School of Earth Sciences & Engineering, Sun Yat-senUniversity, Guangzhou 510275, China,Center for Earth Environment & Resources, Sun Yat-senUniversity, Guangzhou 510275, China;Guangdong Provinical Key Laboratory of Mineral Resources and Geological Processes, Guangzhou 510275, China;School of Earth Sciences & Engineering, Sun Yat-senUniversity, Guangzhou 510275, China and Center for Earth Environment & Resources, Sun Yat-senUniversity, Guangzhou 510275, China;Guangdong Provinical Key Laboratory of Mineral Resources and Geological Processes, Guangzhou 510275, China;School of Earth Sciences & Engineering, Sun Yat-senUniversity, Guangzhou 510275, China |
| |
| Abstract: | Dimensionality reduction, graph data processing, stream data mining, machine learning, association rule algorithm and recommendation system are included in the core technologies of big data and mathematical geoscience. Intelligent geology, including construction of big data-based intelligent metallogenetic and prospecting models, is a highly valuable research direction. Dimensionality reduction aims at extracting low dimensional feature sets out of initial high dimensional feature ones, which can effectively eliminate irrelevant and redundant features, and enhancing the comprehensibility of learning results. Hash algorithm, clustering and PCA are frequently used as tool of dimensionality reduction. Machine learning is the core of artificial intelligence and the fundamental way to endow computer with intelligence. Unity for machine learning and artificial intelligence is emerging. The training model of deep learning often needs huge amounts of data, leading to the raising attention of transfer learning. Graph pattern recognition is an important technology of data mining. Community structure identification has great value to understand the structure and function of the entire network. It can help analyze and predict the interaction between different elements in the network. Immersive virtual reality (VR) technology is another important direction to achieve the visualization of big data. It is of special value in demonstrating mineral resource exploration data characterized by multivariate, heterogeneous, time-spatial, nonlinear, and multi-scale features. Utilizing VR technology to visualize geology and mineral data can result in new insight into mineral exploration under the background of big data era. Infinite data streams widely exist, and even may be automatically and continuously generated in many geological, geochemical, and geophysical monitoring projects. Point query, range query, inner product query, quantile calculation, frequent item-set computing and the like are included in data stream mining. Association rules and recommendation systems, as essential algorithms in data mining, are seeing an expanding application scope. Bayes'' theorem has unique value in the era of big data. The Bayesian Network is a revolutionary tool for genesis modelling. Intelligent Geology (IG) is still at its primary stage. The construction of big data-based intelligent metallogenetic and mineral prospecting models is part of IG. The revolution of research mode of the metallogenetic and mineral prospecting model will emerge with the worldwide participation of teams together with the help of internet and cloud computing technologies. |
| |
| Keywords: | Big Data Mining Dimensionality Reduction Graph Data Processing Infinite Data Stream Machine Learning Association Rule Intelligent Geology Artificial Intelligent metallogenetic Model The Bayesian Network |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《岩石学报》浏览原始摘要信息 |
| 点击此处可从《岩石学报》下载免费的PDF全文 |
|
