Geospatial Data Mining


Data Mining based on geo-referenced data has different characteristics from Data Mining carried out based on business data. Although there is a large number of coincidences between them, there are also some differences, which even though they are not very numerous, are very important and should not be neglected. This course aims to present the Data Mining methodology, as well as its main tools, and also emphasize the specificities that exist in the exploration of georeferenced data. In fact, the fundamental objective is to ensure that students develop their skills to analyze problems based on data, and can use that competence to extract the maximum value from the technologies that we have at our disposal today. Thus, at the end of the course, students must have a good understanding of the main Data Mining tools, as well as a critical spirit regarding their application in the context of geographic information science.

General characterization





Responsible teacher

Fernando José Ferreira Lucas Bação


Weekly - Available soon

Total - Available soon

Teaching language

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


No requirements.


Han, J., Kamber, M. 2006, Data Mining ¿ Concepts and Techniques, Morgan Kaufmann, Elsevier Inc.

Mitchell, T., (1997) Machine Learning, McGraw Hill.

Teaching method

The course is based on self-study, coaching sessions and conducting tutorials and exercises. In the tutorials, students have at their disposal a "script" that allows you to solve the proposed problem, in the case of the exercises is tested student autonomy.
The evaluation includes an examination (40%), a practical project (60%) given on how to report and a presentation.

Evaluation method

Individual Project 80%

Exam 20%

Subject matter

The syllabus is organized in 5 Learning Units (LU):

LU1. Introduction to Data Mining

LU2. Data Mining in the geographic information science context

LU3. The role of Data in Data Mining

LU4. Unsupervised Classification (clustering)

LU5. Supervised Classification (predictive modelling)