On the Impact of Imbalanced Data in Convolutional Neural Networks Performance
In recent years, new proposals have emerged for tackling the classification problem based on Deep Learning (DL) techniques. These proposals have shown good results in certain fields, such as image recognition. However, there are factors that must be analyzed to determine how they influence the results obtained by these new algorithms. In this paper, the classification of imbalanced data with convolutional neural networks (CNNs) is analyzed. To do this, a series of tests will be performed in which the classification of real images of traffic signals by CNNs will be performed based on data with different imbalance levels.
KeywordsDeep learning Convolutional neural network Image recognition Imbalanced dataset
The work of F. Pulgar was supported by the University of Jaén under the Action 15: Predoctoral aids for the encouragement of the doctorate. This work was partially supported by the Spanish Ministry of Science and Technology under project TIN2015-68454-R.
- 2.Kotsiantis, S.: Supervised machine learning: a review of classification techniques. In: Proceedings of the 2007 Conference on Emerging Artificial Intelligence Applications in Computer Engineering: Real Word AI Systems with Applications in eHealth, HCI, Information Retrieval and Pervasive Technologies, pp. 3–24 (2007)Google Scholar
- 8.Zadrozny, B., Elkan, C.: Learning and making decisions when costs and probabilities are both unknown. In: Proceedings of the 7th International Conference on Knowledge Discovery and Data Mining (KDD01), pp. 204–213 (2001)Google Scholar
- 9.Zadrozny, B., Langford, J., Abe, N.: Costsensitive learning by costproportionate example weighting. In: Proceedings of the 3rd IEEE International Conference on Data Mining (ICDM03), pp. 435–442 (2003)Google Scholar
- 11.Goodfellow, I., Bengio, Y., Courville, A.: Deep learning (2016)Google Scholar
- 15.Ciresan, D., Meier, U., Schmidhuber, J.: Multi-column deep neural networks for image classification, Technical Report No. IDSIA-04-12 (2012)Google Scholar
- 16.McMillan, R.L.: How Skype used AI to build its amazing new language translator, wire (2014)Google Scholar
- 17.LeCun, Y., Bengio, Y.: Convolutional networks for images, speech, and time-series. In: Arbib, M.A. (ed.) The Handbook of Brain Theory and Neural Networks (1995)Google Scholar
- 18.Sak, H., Senior, A., Beaufays, F.: Long short-term memory recurrent neural network architectures for large scale acoustic modeling. In: Proceedings of Interspeech, pp. 338–342 (2013)Google Scholar
- 19.Sermanet, P., LeCun, Y.: Traffic sign recognition with multi-scale convolutional networks. In: Proceedings of International Joint Conference on Neural Networks (2011)Google Scholar
- 20.LeCun, Y., Kavukcuoglu, K., Farabet, C.: Convolutional networks and applications in vision. In: Proceedings of 2010 IEEE International Symposium on in Circuits and Systems (ISCAS), pp. 253–256 (2010)Google Scholar