Machine Learning

Objectives

1. Acquire knowledge about data science and machine learning.
2. Using labelled and unlabeled data: supervised classification and regression problems, unsupervised problems.
3. Evaluating the quality of models: using quality measures for regression, classification, and clustering.
4. Know how to use appropriate platforms and libraries for experimentation and problem solving in Machine Learning.

Program

1. Basic concepts about process models, pre-processing, modeling, and current paradigms of Machine Learning (ML).
2. Unsupervised learning algorithms: hierarchical and non-hierarchical clustering.
3. Supervised learning algorithms for regression and classification: decision trees and multilayer perceptron neural networks.
4. Know how to use quality measures per class in classification and global measures for classification and regression. Quality measures for unsupervised models.
5. Use of platforms and libraries for Machine Learning.
6. Solving problems in different learning contexts: image recognition, natural language processing, IoT data processing and robotics.

Teaching Methodologies

The curricular unit will have a theoretical component, where the different forms of learning are presented framed in the techniques of AA addressed and a component of practical activity in which the trainees will have the opportunity to experience and acquire deeper knowledge in each of the topics addressed as well as in the use of libraries and platforms. In asynchronous activities (AA) the trainees solve problems in a collaborative way, using the studied algorithms and concepts.
The final assessment (FA) is made up of an assessment component consisting of asynchronous activities (AA), a group assessment component (TG) in project work format (minimum mark 10) and an assessment test (TA) (minimum mark 9.0).
FA = 0.35 * AA + 0.25 * TG + 0.40 * TA
The students have a positive evaluation if AF is greater than or equal to 9.5. A lower mark implies a written final exame.

Bibliography

Jake VanderPlas. (2016). Python Data Science Handbook. O’Reilly Media. (eBook)
Aurélien Géron. (2019). Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow (2.ª ed.). O’Reilly Media. (disponível em PDF)
Steven L. Franconeri, Lace M. Padilla, Priti Shah, Jeffrey M. Zacks, & Jessica Hullman. (2021). The Science of Visual Data Communication: What Works. Psychological Science in the Public Interest. https://doi.org/10.1177/15291006211051956