Machine Learning is a discipline involving algorithms designed to find patterns in and make predictions about data. Machine Learning is nearly ubiquitous in our world today, and used in things like web analytics, natural sciences and engineering.
This tutorial is intended as an introduction to scikit-learn, a python machine learning package, and to the central concepts of Machine Learning. It is aimed as those with some programming experience, but no practical experiment with machine learning. The workshop will mix important theoretical concepts with hands-on examples. This workshop will cover six main topics
- Data Preprocessing
- Regression Techniques
- Classification Problems
- Cluster Analysis
- Model Selection and Appraisal
- Other topics such as Principal Component Analysis
1 Environment Installation Problem Types (supervised vs unsupervised) Preprocessing - Encoding Categorical Variables Preprocessing - Normalization Creating Simulated Data Training and Testing Data
2 Linear Regression Regression Fitting
3 Binary Logistic Regression SVM Decisions Trees and Random Forests
4 Cluster Analysis Cluster Correctness
5 F-score, Recall Precision Accuracy ROC Curves
6 Curse of Dimensionality PCA