A Robust Learning Model for Dealing with Missing Values in Many-Core Architectures
Most of the classification algorithms (e.g. support vector machines, neural networks) cannot directly handle Missing Values (MV). A common practice is to rely on data pre-processing techniques by using imputation or simply by removing instances and/or features containing MV. This seems inadequate for various reasons: the resulting models do not preserve the uncertainty, these techniques might inject inaccurate values into the learning process, the resulting models are unable to deal with faulty sensors and data in real-world problems is often incomplete. In this paper we look at the Missing Values Problem (MVP) by extending our recently proposed Neural Selective Input Model (NSIM) first, to a novel multi-core architecture implementation and, second, by validating our method in a real-world financial application. The NSIM encompasses different transparent and bound (conceptual) models, according to the multiple combinations of missing attributes. The proposed NSIM is applied to bankruptcy prediction of (healthy and distressed) French companies, yielding much better performance than previous approaches using pre-processing techniques. Moreover, the Graphics Processing Unit (GPU) implementation reduces drastically the time spent in the learning phase, making the NSIM an excellent choice for dealing with the MVP.
KeywordsMissing values Neural Networks GPU
Unable to display preview. Download preview PDF.
- 1.Aikl, L., Zainuddin, Z.: A comparative study of missing value estimation methods: Which method performs better? In: Proc. International Conference on Electronic Design (ICED 2008), pp. 1–5 (2008)Google Scholar
- 2.Ayuyev, V.V., Jupin, J., Harris, P.W., Obradovic, Z.: Dynamic clustering-based estimation of missing values in mixed type data. In: DaWaK 2009: Proceedings of the 11th International Conference on Data Warehousing and Knowledge Discovery, pp. 366–377. Springer, Heidelberg (2009)Google Scholar
- 6.Lopes, N., Ribeiro, B.: Hybrid learning in a multi-neural network architecture. In: Proceedings of the International Joint Conference on Neural Networks (IJCNN 2001), vol. 4, pp. 2788–2793 (2001)Google Scholar
- 8.Lopes, N., Ribeiro, B.: A strategy for dealing with missing values by using selective activation neurons in a multi-topology framework. In: IEEE World Congress on Computational Intelligence, WCCI (2010)Google Scholar
- 9.López-Molina, T., Pérez-Méndez, A., Rivas-Echeverría, F.: Missing values imputation techniques for neural networks patterns. In: ICS 2008: Proceedings of the 12th WSEAS International Conference on Systems, pp. 290–295. World Scientific and Engineering Academy and Society, WSEAS (2008)Google Scholar
- 10.Ribeiro, B., Lopes, N., Silva, C.: High-performance bankruptcy prediction model using graphics processing units. In: IEEE World Congress on Computational Intelligence, WCCI (2010)Google Scholar