# Exploring the Impact of Different Classification Quality Functions in an ACO Algorithm for Learning Neural Network Structures

### Khalid M. Salama, Ashraf M. Abdelbar

#### Abstract

Although artificial neural networks can be a very effective classification method, one of the drawbacks of their use is the need to manually prescribe the neural network topology. Recent work has introduced the ANN-Miner algorithm, an Ant Colony Optimization (ACO) technique for optimizing the topology of arbitrary FFNN's, i.e. FFNN's with multiple hidden layers, layer-skipping connections, and without the requirement of full-connectivity between successive layers. In this paper, we explore the use of several classification quality evaluation functions in ANN-Miner. Our experimental results, using 30 popular benchmark datasets, identify several quality functions that significantly improve on the simple Accuracy quality function that was previously used in ANN-Miner.

#### References

- Boryczka, U. and Kozak, J. (2010). Ant colony decision trees. In 4th International Conference on Computational Collective Intelligence, pages 4373-382.
- Boryczka, U. and Kozak, J. (2011). An adaptive discretization in the ACDT algorithm for continuous attributes. In 3rd International Conference on Computational Collective Intelligence, pages 475-484.
- Dorigo, M. and St ützle, T. (2004). Ant Colony Optimization. MIT Press, Cambridge, MA, USA.
- Liao, T., Socha, K., de Oca, M. M., Stuetzle, T., and Dorigo, M. (2014). Ant colony optimization for mixed-variable optimization problems. To appear in IEEE Transactions on Evolutionary Computation.
- Otero, F., Freitas, A., and Johnson, C. (2009). Handling continuous attributes in ant colony classification algorithms. In IEEE Symposium on Computational Intelligence in Data Mining (CIDM-09), pages 225-231.
- Otero, F., Freitas, A., and Johnson, C. (2013). A new sequential covering strategy for inducing classification rules with ant colony algorithms. IEEE Transactions on Evolutionary Computation, 17(1):64-74.
- Otero, F. E. B., Freitas, A. A., and Johnson, C. G. (2012). Inducing decision trees with an ant colony Applied Soft Computing,
- Parpinelli, R. S., Lopes, H. S., and Freitas, A. A. (2002). Data mining with an ant colony optimization algorithm. IEEE Transactions on Evolutionary Computation, 6(4):321-332.
- Salama, K., Abdelbar, A., and Freitas, A. (2011). Multiple pheromone types and other extensions to the AntMiner classification rule discovery algorithm. Swarm Intelligence, 5(3-4):149-182.
- Salama, K., Abdelbar, A., Otero, F., and Freitas, A. (2013). Utilizing multiple pheromones in an ant-based algorithm for continuous-attribute classification rule discovery. Applied Soft Computing, 13(1):667-675.
- Salama, K. M. and Abdelbar, A. M. (2014). A novel ant colony algorithm for building neural network topologies. In Proceedings ANTS-14. Accepted.
- Socha, K. and Blum, C. (2007). An ant colony optimization algorithm for continuous optimization: Application to feed-forward neural network training. Neural Computing & Applications, 16:235-247.
- Socha, K. and Dorigo, M. (2008). Ant colony optimization for continuous domains. European Journal of Operational Research, 185:1155-1173.

#### Paper Citation

#### in Bibtex Style

@conference{ecta14,

author={Khalid M. Salama and Ashraf M. Abdelbar},

title={Exploring the Impact of Different Classification Quality Functions in an ACO Algorithm for Learning Neural Network Structures},

booktitle={Proceedings of the International Conference on Evolutionary Computation Theory and Applications - Volume 1: ECTA, (IJCCI 2014)},

year={2014},

pages={137-144},

publisher={SciTePress},

organization={INSTICC},

doi={10.5220/0005031301370144},

isbn={978-989-758-052-9},

}

#### in Harvard Style

Salama K. and Abdelbar A. (2014). **Exploring the Impact of Different Classification Quality Functions in an ACO Algorithm for Learning Neural Network Structures** . In *Proceedings of the International Conference on Evolutionary Computation Theory and Applications - Volume 1: ECTA, (IJCCI 2014)* ISBN 978-989-758-052-9, pages 137-144. DOI: 10.5220/0005031301370144

#### in EndNote Style

TY - CONF

JO - Proceedings of the International Conference on Evolutionary Computation Theory and Applications - Volume 1: ECTA, (IJCCI 2014)

TI - Exploring the Impact of Different Classification Quality Functions in an ACO Algorithm for Learning Neural Network Structures

SN - 978-989-758-052-9

AU - Salama K.

AU - Abdelbar A.

PY - 2014

SP - 137

EP - 144

DO - 10.5220/0005031301370144