– Constructive algorithms are effective methods for designing Artificial Neural Networks (ANN) with good accuracy and generalization capability, yet with parsimonious network str...
Leonardo M. Holschuh, Clodoaldo Ap. M. Lima, Ferna...
In this paper the node-level decision unit of a self-learning anomaly detection mechanism for office monitoring with wireless sensor nodes is presented. The node-level decision uni...
A feedforward neural network based on multi-valued neurons is considered in the paper. It is shown that using a traditional feedforward architecture and a high functionality multi-...
Neural-symbolic integration concerns the integration of symbolic and connectionist systems. Distributed knowledge representation is traditionally seen under a purely symbolic pers...
Background: We present a novel method of protein fold decoy discrimination using machine learning, more specifically using neural networks. Here, decoy discrimination is represent...