Predictive Methods of Data Mining

Objectives

The curricular unit of data mining primary goal is to provide students with a fundamental understanding of Predictive Analytics as it relates to improving business performance. This course will cover the basics of predictive analytics and modelling data to determine which algorithms to use and to understand the similarities and differences and which options affect the models most. Topics covered include predictive analytics algorithms for supervised learning, including decision trees, neural networks, k-nearest neighbor, and model ensembles. At the end of the course, participants will be able to use these skills to produce a fully processed data set compatible with building powerful predictive models that can be deployed to increase profitability.

General characterization

Code

200166

Credits

7.5

Responsible teacher

Hours

Weekly - Available soon

Total - Available soon

Teaching language

Portuguese. If there are Erasmus students, classes will be taught in English

Prerequisites

Basic Python knowledge is required. 
Approval on the Curricular Unit of Descriptive methods of Data Mining is advisable.
 

Bibliography

•    Data Mining and Predictive Analytics, Daniel T. Larose, Chantal D. Larose, Wiley, 2015 
•    Applied Predictive Analytics: Principles and Techniques for the Professional Data Analyst, Dean Abbott Wiley, 2014 
•    Machine Learning, Tom M. Mitchell, McGraw Hill, 1997 
•    Pattern Recognition and Machine Learning, Christopher M. Bishop
 

Teaching method

The course is based on a mix of theoretical lectures and practical lectures and tutorials. The theoretical sessions include the presentation of theoretical concepts and methodologies as well as application examples. The main objective of the practical classes is to familiarize students with the software to perform the analysis and data explorations tasks.

Evaluation method

Regular period (1st examination period) 
•    Exam (60%)
•    Project (40%)

Regular period (end examination period)
•    Exam (60%)
•    Project (40%)


Minimum grade of 8.0 (in 20) for the exam and for the project
 

Subject matter

The course is divided in the following learning units (LU): 
LU1.    Introduction to predictive modeling 
LU2.    Data pre-processing 
LU3.    Introduction to classifiers 
a.    Bayesian learning systems 
b.    Instance-based learning and classification 
c.    Regression and classification trees 
d.    Ensemble classifiers 
e.    Neural Networks
LU4.    Predictive models evaluation 
LU5.    Develop a project using Python