2024
Data Processing Technologies in Precision Agriculture
Name: Data Processing Technologies in Precision Agriculture
Code: ERU12710M
3 ECTS
Duration: 15 weeks/84 hours
Scientific Area:
Ciências de Engenharia
Teaching languages: Portuguese
Languages of tutoring support: Portuguese
Sustainable Development Goals
Learning Goals
At the end of this course the student will have acquired knowledge, skills and abilities to:
- Understand a technical report where the information presented result from a statistical analysis and/or
spatial modeling;
- Being able to synthesize information from a qualitative and quantitative dataset, in particular the
interpretation and drawing of conclusions;
- Understand the bi-or multivariate relationships between variables of a dataset, and analyze redundancies
and gaps of information;
- Distinguish the various sub-populations of a sample, and use the best tools for the generation of sub-sets
of data;
- Generate estimated images of a continuous property locally sampled in the study area.
- Apply data analysis tools in R platform, including importing data, graphical view and the output of
reports.
- Understand a technical report where the information presented result from a statistical analysis and/or
spatial modeling;
- Being able to synthesize information from a qualitative and quantitative dataset, in particular the
interpretation and drawing of conclusions;
- Understand the bi-or multivariate relationships between variables of a dataset, and analyze redundancies
and gaps of information;
- Distinguish the various sub-populations of a sample, and use the best tools for the generation of sub-sets
of data;
- Generate estimated images of a continuous property locally sampled in the study area.
- Apply data analysis tools in R platform, including importing data, graphical view and the output of
reports.
Contents
Review of univariate and bivariate statistical analysis and probability theory. Colection of information,
scale, spatial resolution. Graphical representation of data. Uncertainty.
Multivariate analysis. Principal component analysis. Hierarchical and nonhierarchical (K-means)
clustering methods. Analysis of variance (ANOVA).
Regression. Generalized linear models. Spatio-temporal models. Trend curves.
Geostatistics. Random variables. Theory of the regionalized variables. Spatial continuity analysis:
spatial covariance and variogram. Directional variograms and isotropy / anisotropy. Fitting of theoretical
models. Kriging estimation. Kriging variance. View of results. Cross validation.
Pratice: Exercices solved in R software (data analysis and geostatistics).
scale, spatial resolution. Graphical representation of data. Uncertainty.
Multivariate analysis. Principal component analysis. Hierarchical and nonhierarchical (K-means)
clustering methods. Analysis of variance (ANOVA).
Regression. Generalized linear models. Spatio-temporal models. Trend curves.
Geostatistics. Random variables. Theory of the regionalized variables. Spatial continuity analysis:
spatial covariance and variogram. Directional variograms and isotropy / anisotropy. Fitting of theoretical
models. Kriging estimation. Kriging variance. View of results. Cross validation.
Pratice: Exercices solved in R software (data analysis and geostatistics).
Teaching Methods
The teaching methodology uses theoretical and practical sessions of 1 and 2 hours respectively:
i)
theoretical lectures with powerpoint
ii) practical classes in the computer room. The theoretical
explanations are suported with practical examples related with the master course. The classes are based
on problem solving, taking as a starting point realistic datasets that reproduce some of the situations that
future professionals will work.
The evaluation is preferably of continuous type but altrnatively can be made by classical exam. Two
written tests for methods (representing 25% + 25% of the final grade), and a report made by groups of
two students with a resolution of the practical problems worked in practices (remaining 50 %) will
be developed for the continuous assessment option. Alternatively, students have a final theoretical
examination where the grade of the theoretical component tests can be also improved
i)
theoretical lectures with powerpoint
ii) practical classes in the computer room. The theoretical
explanations are suported with practical examples related with the master course. The classes are based
on problem solving, taking as a starting point realistic datasets that reproduce some of the situations that
future professionals will work.
The evaluation is preferably of continuous type but altrnatively can be made by classical exam. Two
written tests for methods (representing 25% + 25% of the final grade), and a report made by groups of
two students with a resolution of the practical problems worked in practices (remaining 50 %) will
be developed for the continuous assessment option. Alternatively, students have a final theoretical
examination where the grade of the theoretical component tests can be also improved
