Maharaj, Rashin and Bheekun, Shaan (2004) Abalone Harvest Prediction using AI methods, CS04-15-00, Department of Computer Science, University of Cape Town.
![[img]](http://pubs.cs.uct.ac.za/style/images/fileicons/application_pdf.png) |
PDF
Abalone_Harvest_Prediction_using_AI_methods.pdf Download (256kB) |
Abstract
“Foreknowledge of the future makes it possible to manipulate both enemies and supporters." --Raymond Aron in “The Opium of the Intellectuals” The above quote describes perfectly the motivation for developing a prediction tool. Being able to minimise uncertainty to any possible degree will give any business that engages in prediction or forecasting, a competitive advantage that is becoming necessary for economic prosperity. The main task of this project was to use Artificial Intelligence Methods to support Abalone Harvest Prediction. The first step was to choose a suitable graphical probabilistic network so that the abalone growth, given the factors that affect it, can be successfully modelled. An implementation structure for the chosen model was designed and then implemented. Once a suitable model was designed, the two core components of the system were implemented. A learning engine to learn the parameter values for the chosen model, and an inference engine to perform probabilistic inference on the learnt parameters. A graphical user interface that is user-friendly and easy to understand by the people at the farm was then developed and implemented. This graphical user interface hides the complexities of artificial intelligences techniques of which can intimidate the novice user.
| Item Type: | Technical report |
|---|---|
| Uncontrolled Keywords: | Artificial Intelligence, Bayesian Network, Abalone |
| Subjects: | Computing methodologies > Artificial intelligence |
| Date Deposited: | 21 Oct 2004 |
| Last Modified: | 10 Oct 2019 15:35 |
| URI: | http://pubs.cs.uct.ac.za/id/eprint/171 |
Actions (login required)
 |
View Item |