Data Analysis, Optimization and Decision Making
Code: 44733 ECTS Credits: 6| Degree | Type | Year |
|---|---|---|
| 4318303 Reseach and Innovation in Computer Based Science and Engineering | OT | 0 |
Contact
- Name:
- Ramon Vilanova Arbos
- Email:
- ramon.vilanova@uab.cat
Teachers
- Antoni Morell Perez
- Jose Lopez Vicario
- Carles Pedret Ferré
- Carlos Sanchez Ramos
Teaching groups languages
You can view this information at the end of this document.
Prerequisites
N/A
Objectives and Contextualisation
The main objective of this subject is for the student to be able to understand what is the best possible strategy to optimize the treatment of the data to be analyzed. To do this, different techniques will be presented to process the input data (Time Series Analysis, coding in SVM or Random Forest or, in terms of text processing, techniques such as the Bag of Words or LDA). In a more advanced way, the use of techniques such as genetic algorithms or neural networks will be explored. In the optimization part, linear and non-linear methods will be studied, in addition to covering multi-objective optimization methodologies. Finally, advanced decision-making concepts will be introduced, touching on aspects such as the introduction of risk and uncertainty associated with the information to be analyzed.
Learning Outcomes
- CA10 (Competence) Design the correct data processing strategy to obtain the expected result.
- CA11 (Competence) Propose a robust decision-making system that considers the associated risk and uncertainty given the available information.
- KA14 (Knowledge) Describe the most appropriate data representation techniques for solving a specific problem.
- KA15 (Knowledge) Identify appropriate optimisation and decision-making strategies in order to comply with the restrictions of the problem at hand, and obtain optimum algorithm performance.
- SA18 (Skill) Apply the most appropriate data manipulation and representation techniques to the problem at hand.
- SA19 (Skill) Critically apply optimisation methods to decision-making processes for scientific and engineering problems.
- SA20 (Skill) Formulate decision-making problems so that they can be tackled using numerical methods and offer solutions that incorporate considerations related to risk and uncertainty.
Content
Exploratory data analysis
- Introduction to Data Processing. Main aplication areas and problems
- Data representation, feature extraction.
- Data structure exploration, visualization and clustering
- Dimensionality reduction and feature selection.
- Supervised Methods for Data Analysis: SVM, regression.
- Validation. Metrics, analisis of bias in models, statistical tools, trustworthiness
Optimization
- linear programming
- Non-linear optimisation
- Duality, multipliers, dynamic programming
- modelling and optimisation software
multi-objective optimisation
- Multicriteria decisión making
- Methods
- Multicriteria, preferences
- Uncertainty and risk
Advanced topics and applications
Activities and Methodology
| Title | Hours | ECTS | Learning Outcomes |
|---|---|---|---|
| Type: Directed | |||
| Face-to-face classroom | 30 | 1.2 | |
| Type: Supervised | |||
| Supervised Activity | 15 | 0.6 | |
| Type: Autonomous | |||
| Autonomous activity | 90 | 3.6 |
This subject has a marked engineer character. Theory: it is rather a methodology, therefore trying to promote methodological application instead of theoretical developments. At the end of the subject, assignments/projects will be presented for evaluation.
Annotation: Within the schedule set by the centre or degree programme, 15 minutes of one class will be reserved for students to evaluate their lecturers and their courses or modules through questionnaires.
Assessment
Continous Assessment Activities
| Title | Weighting | Hours | ECTS | Learning Outcomes |
|---|---|---|---|---|
| Project and exercises | 100 | 15 | 0.6 | CA10, CA11, KA14, KA15, SA18, SA19, SA20 |
This subject is assessed on the basis of a work/project: in which you will have to deal with a problem based on the elements seen during the subject. A report must be presented and a presentation made.
Bibliography
Reference material and sources will be provided in each section
Software
MATLAB
Language list
| Name | Group | Language | Semester | Turn |
|---|---|---|---|---|
| (PLABm) Practical laboratories (master) | 1 | English | first semester | afternoon |
| (TEm) Theory (master) | 1 | English | first semester | afternoon |