ANALYSIS OF ENVIRONMENTAL DATA 
WITH MACHINE LEARNING METHODS

Location: Jozef Stefan Institute, Jamova cesta 39, Ljubljana

This is the twelfth seminar on the above topic since November 1996.
The last edition of this course (March 2008) was significantly extended as compared to its previous edition (February/March 2006) to extensively cover many different application areas and case studies.
An online version of the introductory talk at a previous seminar of this series can be found here.
Information on previous seminars in this series can be found here.

Information on travel to Slovenia and Ljubljana can be found here.

in cooperation with

Jozef Stefan International Postgraduate School and
University Nova Gorica, School of Environmental Sciences

For graduate students of institutions that participate in the ECTS, including the Jozef Stefan International Postgraduate School and the School of Environmental Sciences, University of Nova Gorica, the seminar counts as regular coursework.

The seminar is part of and supported by the EU funded project TEMPUS CD-JEP-41038-2006 Ecosystem Informatics: Development of a Postgraduate Curriculum. The seminar is also supported by the EU funded FP7 project AgroSense.

Environmental data often need to be analysed in order to obtain information necessary for environmental management decisions. Sometimes it is necessary to use the data to build a model of the environmental process that we want to manage. In other cases the analysis is necessary to understand the environmental processes studied by identifying and understanding the interrelationships of different parameters.

Machine learning can be used to elicit regularities from data. In comparison to simple forms of regularities/dependencies treated by statistical methods, machine learning methods can find more complex regularities/dependencies that include both numerical and logical conditions.

The seminar will give an introduction to selected machine learning methods as well as illustrative case studies of using these methods to analyse environmental data. Applications in the areas of aquatic ecosystems, agriculture, forestry, environmental epidemiology, and disaster forecasting/relief will be covered in detail. The participants will learn to use selected machine learning tools and will have the opportunity for practical work with these tools on real environmental data.

The seminar is intended for researchers and other professionals whose work requires the analysis of environmental data or modeling of ecosystems and ecological processes. The seminar is intended for an audience with a diverse background, and has been in the past attended by particpants coming from the areas of biology, chemistry, environmental science, and other areas related to ecology and environmental management, as well as computer science and information technology, The machine learning methods and tools introduced are applicable to data analysis problems from different areas of environmental science and management, as well as data analysis problems from other areas.

Contents

• Machine learning methods and methodology
• Machine learning, data mining and knowledge discovery
• Decision trees
• Classification rules
• Naive Bayes
• Nearest neighbor
• Evaluating classifiers
• Ensemble methods
• Feature weighting, ranking, and selection
• Equation discovery
• Automated modeling of dynamic systems
• Application areas
• Aquatic ecosystems
• Agriculture
• Forestry
• Environmental epidemiology
• Disater forecasting and relief
• Case studies of using machine learning to analyse ecological data
• Modeling algal growth in lakes and lagoons
• Modeling gene-flow between GM and conventional crops
• Predicting forest properties using remote sensing data
• Investigating the effects of mercury on miners
• Predicting danger of fire in the natural environment
• ... and many more
• Demonstrations/hands-on exercises/practical work
  with a machine learning software package on real ecological data and
  individual consultations with lecturers
• Participant presentations and discussion

A detailed programme/schedule can be found here.

---

---

Duration:

Lecturers

• Academician Professor Ivan Bratko, Faculty of computer and information sciences, University of Ljubljana and Jozef Stefan Institute, Ljubljana
• Professor Saso Dzeroski, Jozef Stefan Institute, Ljubljana, and Jozef Stefan International Postgraduate School
• Professor Boris Kompare, Faculty of civil and geodetic engineering, University of Ljubljana
• Professor Marko Debeljak, Jozef Stefan Institute, Ljubljana
• Dr. Ljupco Todorovski, Faculty of administration, University of Ljubljana
• Dr. Natasa Atanasova, Faculty of civil and geodetic engineering, University of Ljubljana
• Dr. Bernard Zenko, Jozef Stefan Institute, Ljubljana
• Andrej Kobler, MSc., Slovenian Forestry Institute, Ljubljana

Please register by the 10th of April at the latest. The number of participants is limited, so early registration is recommended. Registration is on a first-come first-served basis. The registration fee for the seminar is 800 EUR. It includes access to the lectures, demonstrations and hands-on work, seminar materials, lunches, refreshments, one social dinner and VAT (DDV). Graduate students funded by the Slovenian Research Agency under the young researchers programme (mladi raziskovalci) are eligible for a 25% discount.

Payment should be done by the 15th of April by bank transfer or check. Please send receipt of payment or check made out to:

          Account holder: Jozef Stefan Institute,
                          Jamova cesta 39, 1000 Ljubljana, Slovenia
          Account number: 01100-6030344242
               Bank Name: BANK OF SLOVENIA 
            Bank Address: Slovenska 35, 1505 Ljubljana, Slovenia
                    IBAN: SI56011006030344242
         S.W.I.F.T. code: BSLJSI2X
   Reference for payment: AEDML COURSE 0308

Name:         

Institution:  

Address:      

Position:     

Phone, fax:  

E - mail: