A comparative study of the IDS method and feedforward neural networks

The ink drop spread (IDS) method is a modeling technique which has been used in a learning methodology called the active learning method (ALM). The IDS method is characterized by intuitive pattern-based processing and the architecture comprising heavily parallelized processing units. While being analogous to neural networks in structural characteristics, the IDS method does not require intricate calculations and iteration of the same training data set observed in the learning of neural networks. This paper describes a comparative study of the IDS method and the standard feedforward neural network in terms of their algorithmic and architectural characteristics and shows the effectiveness of the IDS method through regression modeling and classification tasks.