FCUP-CC4018-2024/2025-1S: Arbeitsmaterial

Abschnitt	Name	Beschreibung
	UC info
	Self-assessment #1
	Introduction to Data Science (in Portuguese)

Assignment	DATASET
	Guideline for diagnostic (in Portuguese)
	Table with BMI information for Portugal (in Portuguese)
	Explanation about the PPA variable (in Portuguese)
	Tables of blood pressure for children (in English)
	Paper #1 with results on a similar dataset (169 subjects)
	Paper #2 with statistical results on the same dataset (7199 subjects)
	Data Analysis: some good practices

Homeworks	To read #1: Apples-to-apples: the pitfals of cross-validation
	To read #2: The relationship between ROC and Precision-Recall curves
	To read #3: refutation of the second paper

Python Books	Python Data Science Handbook
	Suggestions from python.org
	O'Reilly Python books
	Python resources

Theoretical Classes	Presentation
	Introduction to Data Mining
	Data understanding and manipulation
	Distances and dimensionality reduction (till slide 41, inclusive)
	Recorded class
	Distances and dimensionality reduction (cont. from slide 42)
	Distances and dimensionality reduction (cont. from slide 55)
	Data imputation
	Data visualization
	Basic Concepts in Classification
	Basic Concepts in Classification: Decision Trees (from slide 15)
	Naive Bayes classifier
	Naive Bayes classifier (from slide 9) and Belief Networks
	Evaluating the Performance of a Classifier
	Evaluation Metrics
	Evaluating the Performance of a Classifier (from slide 10)
	Evaluation Metrics (revisited)
	Regression and KNN
	Python code associated with the regression and KNN slides
	Melbourne data associated with regression and KNN slides
	Support Vector Machines (SVM)
	A little more detail on SVMs	Section 2.6.1.4 of this dissertation has a detailed and nice explanation about SVMs.
	Artificial Neural Networls
	Clustering
	Ensembles
	Basic Association Analysis

Practical Classes	Entropy revisited
	Distances revisited
	species.csv
	Code for decision trees (iris, with pruning)
	Code for naive Bayes (German credit dataset)
	Decision boundaries
	Performance Evaluation of Classifiers
	Regression and Logistic Regression
	SVM exercises
	Hierarchical Clustering	Here it is some Python code applying hierarchical clustering to the iris dataset. Explore the various options of clustering, including k-means, k-means++ and dbscan. Identify differences between these different clustering methods. Apply these methods and evaluate the quality of the generated clusters using your favorite dataset.
	PPT_to_PDF	PPT_to_PDF_DIR