1. To know the fundamental concepts and methodologies of Data Science.
2. To know how to explore data and build graphical visualization.
3. To know and learn to apply simple methods of statistical learning.
4. To create and use models of decision trees and bayesian classification.
5. To use methods for supervised models evaluation.
6. To acquire competences in the use of R and Python languages in Data Science applications.
"Part I introduction to data and the science
1. Decision Support Systems (DSS). The CRISP-DM methodology. Big data and Data Miing technologies.
2. Information pre-processing. ETL and data cleansing. Visualization and data exploration.
3. Introduction to machine learning models and methods. Supervised and unsupervised problems. Model assessment and validation. Consolidation, dissemination, and implementation of extracted knowledge."
"Part II Unsupervised and simple supervised methods
1. Unsupervised classification algorithms as cluster analysis.
2. Simple supervised algorithms. Naïve Bayes and belief networks.
3. Use and interpret assessment measures from supervised and unsupervised models."
"Part III Learning Trees and Rules
1. Classification and regression trees induction.
2. Supervised algorithms for rule induction and association rules.
3. Pruning and interpretation of results. The overfitting problem."
"Part IV Learning with Artificial Neural Networks
1. The Perceptron Algorithm. Multilayer networks.
2. The Perceptron Training Rule, descending gradient criterion.
3. Introduction to Deep Learning, Convolutional and Residual Neural Networks."
In the lecture sessions in-person teaching methodologies are applied with an expository and demonstrative method, using visual aids and examples.
In the laboratory sessions, students solving consolidation problems based on descriptions and accompanied by the teacher, allowing them to work autonomously but under supervision.
In the project, students are challenged to identify data in any format and on a topic chosen by the group, that will be worked on by the algorithms studied. The results are analysed and interpreted.
Problems are solved by students on a weekly basis and follow-up reports are discussed.
All the support materials are made available on Moodle, and this tool is also used to deliver and discuss the projects.
Essencial
Complementary
01060999
6