Training Cost-Sensitive Neural Networks with Methods Addressing the Class Imbalance Problem
IEEE Transactions on Knowledge and Data Engineering
Classification error of multilayer perceptron neural networks
Neural Computing and Applications
IEEE Transactions on Knowledge and Data Engineering
Hi-index | 0.00 |
A challenge in systems biology is to discover new functional associations between proteins and, from these new relationships, assign functions to unannotated proteins. High-throughput data can be the key to achieve this task, but they often lack the degree of specificity needed for predicting accurate protein functional associations. This improvement in specificity can be achieved through the integration of heterogeneous data sets in a suitable manner. In this paper, we assess the quality of prediction of functional associations between proteins in two cases: (1) using each data source in isolation and (2) by means of Artificial Neural Networks through the integration of different biological data sources.