Teacher: Professor Rahim Barzegar, Institut de recherche en mines et en environnement (IRME)
 Rahim.Barzegar@uqat.ca

This is an online course
The course aims to:

• Learn and apply fundamental statistical methods to interpret geological data sets, including descriptive statistics, hypothesis testing and data visualization.
• Develop proficiency in multivariate statistical techniques such as clustering, principal component analysis (PCA) and correlation analysis understand complex relationships within geological data.
• Apply machine learning methods for regression and classification in the geosciences, learning to implement models such as linear regression, decision trees and ensemble methods.
• Acquire practical skills using Python libraries such as Pandas, NumPy, SciPy, Scikit-learn and Matplotlib to analyze geoscientific data.
• Carry out real-life case studies involving the analysis of spatial and temporal data identify trends and patterns in geological datasets.
By the end of the course, students will be able to carry out data-driven assessments in the geosciences, drawing on theoretical and practical knowledge to address real-world challenges.
Introduction
− Reading and adoption of the lesson plan
− Introduction to geoscientific data types
− Basics of statistics and machine learning
Data description and uncertainty
− Descriptive statistics: mean, median, mode
− Data visualization: histograms, normality tests, outliers, modality, box and whisker plots, stem and leaf plots
− Sources uncertainty in geological data
− Error propagation data analysis
Handling missing values
− Strategies for dealing with missing values
− Imputation techniques and data integrity
Statistical tests
− Confidence levels, value and rank tests
− Introduction to Z, t, Chi-square, Kolmogorov-Smirnov, and Mann-Whitney tests
− Practical work #1
Multivariate techniques
− Sum-of-squares methodology
− Principal component analysis (PCA) and factor analysis
− Cluster analysis: K-means, hierarchical clustering
Spatial data analysis
− Spatial distribution of data
− Semi-variograms
− Interpolation
− Practical work #2
Practical session-1 (Lab)
− Introduction to the SPSS interface
− Overview of data entry, variable and SPSS navigation
− Practical exercises on data description and statistical tests
− Practical exercises on multivariate techniques: PCA, factor analysis and cluster analysis in SPSS
− Spatial data visualization and basic interpolation with SPSS
Analysis of temporal data
− Time series data bases in geosciences
− Time trend analysis, seasonality and cycle detection
− Time series decomposition, smoothing techniques and autocorrelation
− Applications temporal data analysis for monitoring environmental change
Introduction supervised and unsupervised learning
− Overview of supervised vs. unsupervised learning, applications in geosciences
− Key concepts: training data, labels, clustering, dimension reduction
− Introduction to Python libraries for machine learning: Scikit-learn, TensorFlow
Supervised learning: Regression and classification techniques
− Curve fitting, linear and non-linear regression models
− Regression models: neural networks, decision , set models
− Classification models: k-Nearest Neighbors (KNN), decision trees, support vector machines (SVM)
− Model evaluation
Unsupervised learning: Clustering and dimension reduction
− DBSCAN (Density-based clustering noisy applications)
− Self-organizing maps (SOMs)
− Uniform Manifold Approximation and Projection (UMAP)
− Practical work #3
Practical session-2 (Lab)
− Practical exercises on regression, classification and clustering problems in the geosciences