| Abstract
| - Self-organizing maps (SOMs) are a type of artificial neural network that through training can producesimplified representations of large, high dimensional data sets. These representations are typically used forvisualization, classification, and clustering and have been successfully applied to a variety of problems inthe pharmaceutical and bioinformatics domains. SOMs in these domains have generally been restricted tostatic sets of nodes connected in either a grid or hexagonal connectivity and planar or toroidal topologies.We investigate the impact of connectivity and topology on SOM performance, and experiments wereperformed on fixed and growing SOMs. Three synthetic and two relevant data sets from the chemistrydomain were used for evaluation, and performance was assessed on the basis of topological and quantizationerrors after equivalent training periods. Although we found that all SOMs were roughly comparable atquantizing a data space, there was wide variation in the ability to capture its underlying structure, andgrowing SOMs consistently outperformed their static counterparts in regards to topological errors.Additionally, one growing SOM, the Neural Gas, was found to be far more capable of capturing details ofa target data space, finding lower dimensional relationships hidden within higher dimensional representations.
|
