Applied Environmental Informatics
Type of course in situ
Course provider Cranfield University
Available dates for this course
Please contact Cranfield University directly for a current list of course dates
A basis and understanding of methods pertaining to Informatics is needed to effectively obtain information from data. The objective of the module is to supply the student with a toolbox of techniques for data mining and modelling (informatics) and develop in the student the strategic ability to effectively apply this toolbox.
Introduction to computational methods in informatics; How do we turn data into information?
Strategies and approaches to manage large data for computational analysis.
Data exploration and data mining. Strategies to elucidate underlying structures in the data. Are these causal or coincidental? How does one interpret and communicate results from a data mining exercise effectively.
Inference modelling. Generating quantitative models which can be deployed on existing or new data to generate the required information. Identifying the appropriate model type, form and configuration. Developing the technical skill to configure and deploy inference engines.
Process, empirical or semi-empirical modelling. To use existing process based or (semi) empirical models to generate information from data. Identifying and understanding the constraints of each type of model. Developing the technical skill to configure and deploy these models. Introduction to model data fusion methods and their applications.
Understanding how models operate in space and over time and how spatial and/or temporal effects can affect model behaviour.
Error and Diagnostics. How to assess model performance. Validation procedures and measures. Sources of uncertainty (data/model/deployment). Using model performance measures as diagnostic tools for optimal model configuration and on-going quality control.
Effective communication of the computational Informatics processes and outcomes. Statement of quality and remit of the modelling development process. Identifying and communicating what a particular model can and cannot do.
On successful completion of this study the student should be able to:
1. Assess the potential and potential pitfalls of Ã¢â€šÂ¬Ã‹Å“big dataÃ¢â€šÂ¬Ã¢â€žÂ¢.
2. Assemble and organize data for prescribed analysis and modelling approaches.
3. Appraise and apply data mining techniques, identify underlying data structures.
4. Construct models that reproduce observed relationships; the application of inference engines.
5. Create integrative designs of process models with data; applying model data fusion.
6. Recognize uncertainty and error in data and model parameter estimations.
7. Develop diagnostics measures of model performance.
To book on this course, find out the dates please contact:
T: +44 (0) 1234 754176
F: +44 (0) 1234 751206
Cranfield Campus College Road Cranfield MK43 0AL UK
Telephone: +44 (0)1234 758540