第一章:绪论 (2 学时)
讲授 科学第四范式、地球科学大数据机器挖掘、建模
第二章:数据清洗与预处理(6 学时)
讲授 数据清洗的概念、数据集成与融合、数据变换、数据规约、离群点检测
第三章:高维数据降纬(6 学时)
讲授 相关分析、典型相关分析、哈希算法、主成分分析、因子分析
第四章:分类与预测(6 学时)
讲授 回归分析、聚类分析、判别分析、关联规则算法、推荐系统算法
第五章:无限流数据与时间序列(6 学时)
讲授 无限流数据与时序模式、数据特征提取、时间序列算法
第六章:机器学习与深度学习(10 学时)
讲授 机器学习的发展史、机器学习分类、人工神经网络、深度学习
第七章:应用实践:利用机器学习区分地震信号与背景噪声(12 学时)
实验 利用 Python 库 TensorFlow,Keras 等进行地震观测值样本训练,评价不同分类方法表现
Chapter 1: Introduction (2 hours)
The fourth normal form of science, big data machine mining and modeling of earth sciences
Chapter 2: Data Cleaning and Preprocessing (6 hours)
The concepts of data cleaning, data integration and fusion, data transformation, data protocol, outlier detection
Chapter 3: Dimension Reduction in High-Dimensional Data (6 hours)
Correlation analysis, canonical correlation analysis, hash algorithm, principal component analysis, factor analysis
Chapter 4: Classification and Forecasting (6 hours)
Regression analysis, cluster analysis, discriminant analysis, association rule algorithm, recommendation system
algorithm
Chapter 5: Infinite Stream Data and Time Series (6 hours)
Infinite stream data and time series patterns, data feature extraction, time series algorithms
Chapter 6: Machine Learning and Deep Learning (10 hours)
The history of machine learning, classification of machine science, artificial neural networks, and deep learning
Chapter 7: Practise: Seismic signal/noise discrimination with machine learning (12 hours)
Using Python modules such as TensorFlow, Keras to apply machine learning to train dasets, assessing the performace
of different classifiers
教材:
周永章,张良均,张奥多,王俊。地球科学大数据挖掘与机器学习,中山大学出版社,2018
参考资料:
1. Trevor Hastie, Robert Tibshirani, Jerome Friedman. The Elements of Statistical Learning, Springer, 2017
2. Ian Goodfellow,Yoshua Bengio,Aaron Courville. Deep Learning, MIT Press, 2016
3. 周志华。机器学习,清华大学出版社,2016