Machine Learning I
Lecture notes
_____________________________________________________________________________________
Introduction to Machine Learning
1. Guide (PDF)
1.1 Presentation (PDF)
The subject is introduced by illustrating practical applications and explaining the three main concepts: tasks, models, and algorithms.
Models: KNN and decision trees
2. Guide (PDF)
2.1 Presentation (PDF)
Two simple but powerful machine learning models are introduced: K-nearest neighbour and decision trees, both for classification and regression.
Model evaluation
3. Guide (PDF)
3.1 Presentation (PDF)
The importance of evaluating models is discussed, and also several methods and measures for model evaluation.
Large Scale Machine Learning: MapReduce and Spark
4. Guide (PDF)
4.1. MapReduce Presentation (PDF)
MapReduce is introduced as a data parallelism programming model.
4.2. Spark Presentation (PDF)
Spark is introduced as a data parallelism programming model that solves some issues of MapReduce.
Attribute selection and transformation
5. Guide (PDF)
5.1 Presentation (PDF)
This lecture explains methods for selecting (filter, wrapper, ...) and transforming attributes (PCA, Random Projections).
Ensembles
6. Guide (PDF)
6.1 Presentation (PDF)
This lecture explains how to improve models by means of ensemble techniques (Bagging and Boosting).