Knowledge Extraction and Advanced Analyses Through Inverse Modelling Using Artificial Neural Networks

Wolfram C Rinke, Albert F Stöckl and Andreas B Eisingerich
Additional contact information
Wolfram C Rinke: Fachhochschule Burgenland GmbH, Austria
Albert F Stöckl: Fachhochschule Krems, Austria
Andreas B Eisingerich: Imperial College UK, UK

from ToKnowPress

Abstract: Using artificial neural networks (ANN) to model an observed process is state of the art in engineering and is receiving more and more attention in social or marketing research. As shown in previous publications static data analytics, like an ANN based dependency matrix (DM), creates a better understanding of the relationship between dependant and independent variables of the observed system. To gain a better understanding of the behaviour and dynamics of an observed system a further step has been taken. . This is accomplished by transforming the ANN-DM into its open form equivalent. This is represented by several ANN models, one for each observed model parameter. An algorithm, which was published by the author, can be used for this transformation. The final result makes it possible to study the dynamic relationships between all parameters, including simulation and conclusions on its inverted model behaviour. The author will show based on an example from tourism marketing, how this inverse modelling approach can be applied and what new knowledge can be extracted from the achieved simulation results. The conclusion is that ANN based algorithms makes it possible to model an observed system in a static but also in a dynamic way. Inverting the model generates a deeper insight view and additional knowledge about an observed system. The resulting applications range from supporting strategic decisions to predictive control or model based simulation.

Keywords: knowledge extraction; model inversion; artificial neural networks; decision support systems; simulation (search for similar items in EconPapers)
Date: 2016
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.toknowpress.net/ISBN/978-961-6914-16-1/papers/ML16-099.pdf full text (application/pdf)
http://www.toknowpress.net/ISBN/978-961-6914-16-1/MakeLearn2016.pdf Conference Programme (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:tkp:mklp16:503-511

Access Statistics for this chapter

More chapters in Managing Innovation and Diversity in Knowledge Society Through Turbulent Time: Proceedings of the MakeLearn and TIIM Joint International Conference 2016 from ToKnowPress
Bibliographic data for series maintained by Maks Jezovnik ().