Exploratory Data Analysis and Data Preprocessing
- Introduction to Data Analytics
- Exploratory Data Analysis (EDA) (Visualization, and Descriptive Statistics)
- Hands-on EDA
- Data Preprocessing (including PCA)
- Hands-on Data Preprocessing
Objective: On the first day of the course, the participants will be able to get a bird-eye view of the data analytics process. They will explore two of the four data analytics modules called exploratory data analysis, a necessary step to get the feel of the data, and data preprocessing, a necessary step to clean and format the data before building machine learning models. Learning will be reinforced via hands-on training in R.
Supervised Machine Learning
- Decision Tree
- Hands-on Decision Tree
- Regression (Linear and Logistics)
- Hands-on Regression
- Model Evaluation
Objective: On the second day of the course, the participants will be able to build data-driven models via supervised machine learning algorithms. Two representative machine learning algorithms - one for classification and the other for regression - will be covered. Model evaluation matrices (e.g., confusion matrix, ROC, AUC, etc.) and model evaluation methods (e.g., 10-fold cross validation) will be discussed towards the end of the day.
Ensemble Methods and Unsupervised Machine Learning
- Hands-on Model Evaluation
- Ensemble Methods (Bagging, Boosting and Random Forest)
- Hands-on Ensemble Methods
- Cluster Analysis (k-Means and Hierarchical)
- Hands-on Cluster Analysis
- Class feedback and wrap up
Objective: On the third day of the course, the participants will be introduced to more advanced machine learning techniques of ensemble methods and unsupervised machine learning algorithms. They will be able to implement the complete data mining pipeline including model building and model evaluation in R.
Geoscientists, Engineers, IT professionals and aspiring Citizen Data Scientists working in the oil and gas industry who want to get introduced to data analytics techniques for building data-driven models.