Computational Power of Killers and Helpers in the Immune System
The natural immune system is a subject of great research interest because of its powerful information processing capabilities. It uses characteristics such as learning, memory and associative retrieval to solve recognition and classification tasks. The model presented in this work belongs to the cla...
| Main Author: | |
|---|---|
| Format: | masterThesis |
| Language: | por |
| Published: |
2009
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10451/13958 |
| Country: | Portugal |
| Oai: | oai:repositorio.ul.pt:10451/13958 |
| Summary: | The natural immune system is a subject of great research interest because of its powerful information processing capabilities. It uses characteristics such as learning, memory and associative retrieval to solve recognition and classification tasks. The model presented in this work belongs to the class of models introduced by Farmer et al and is inspired by the hypothesis of clonal selection theory and idiotypic network introduced by Niels Jerne. The main objective is to present a modified Immunological Algorithm that can be used in order to solve problems much in the way that Evolutionary Algorithms or certain types of Artificial Neural Networks do. Besides presenting the algorithm itself we discuss his various parameters, the way to present problems to it and how to extract results from its outcome. The model is then described as being a meta-algorithm to the Probabilistic Algorithms set. Several real problems are presented in order to compare this model with other types of Biologically inspired solving problems models. Finally we discuss various metrics to compare the efficiency and the results of the various biologically inspired models |
|---|