Similarity networks for classification: a case study in the Horse Colic problem
This work addresses classification challenges in veterinary medicine by proposing an incremental method for handling mixed data types in neural networks.
The paper tackles the Horse Colic classification problem by developing a two-layer neural network that uses a similarity-based neuron model to handle diverse data types and missing values, achieving results comparable to radial basis function networks in this specific dataset.
This paper develops a two-layer neural network in which the neuron model computes a user-defined similarity function between inputs and weights. The neuron transfer function is formed by composition of an adapted logistic function with the mean of the partial input-weight similarities. The resulting neuron model is capable of dealing directly with variables of potentially different nature (continuous, fuzzy, ordinal, categorical). There is also provision for missing values. The network is trained using a two-stage procedure very similar to that used to train a radial basis function (RBF) neural network. The network is compared to two types of RBF networks in a non-trivial dataset: the Horse Colic problem, taken as a case study and analyzed in detail.