Publication in the Diário da República: Despacho n.º 11262/2016 - 19/09/2016
8 ECTS; 1º Ano, Anual, 36,0 PL + 8,0 TP + 4,0 S + 32,0 OT + 10,0 O
Lecturer
- Sandra Maria Gonçalves Vilas Boas Jardim
Prerequisites
Not applicable.
Objectives
A. Be able to extract the data to be integrated into the information/knowledge pyramid,
B. Be able to apply data pre-processing techniques
C. Know the main analytic models and the algorithms that implement them
D. Design and implement algorithms for the creation of analytical models
Program
1. Data acquisition and pre-processing (data quality, aggregation, sampling, dimensionality reduction, feature subset selection, feature creation, discretization and binarization, attribute transformation)
2. Analytics techniques: forecasting techniques, descriptive analytics (clustering technique), predictive analytics (machine learning, regression and text mining), decision optimization techniques
3. Algorithms: algorithms for implementing the techniques discussed in point 2.)
4. Visualization and interpretation of results from analytics systems (dashboards)
Evaluation Methodology
Continuous assessment:
Written test (30%), Weekly individual tasks (30%), Final practical assignment (40%)
Exam-based assessment (First attempt and Resit)
Written test (50%), Practical Assignment (50%)
In all evaluation periods, the minimum grade for each component is 7 values.
In each evaluation period, approval to the course implies a final classification equal to or greater than 10 values.
Bibliography
- Turban, E. e Sharda, R. e Dursun, D. e King, D. (2011). Business Intelligence: a managerial approach. New Jersey: Prentice Hall
- Howson, C. (2008). Successful business intelligence : secrets to making BI a killer app.. New York: Mc Graw-Hill
- Eckerson, W. (2006). Performance Dashboards: Measuring, Monitoring, and Managing Your Business. New Jersey: John Wiley & Sons
Teaching Method
Lectures and practical sessions.
Software used in class
Keras, Tensorflow, productivity tools Elearning platform.
