Application of self-organizing feature maps with lateral inhibition to structure a library of reusable software components

This paper is concerned with the application of Kohonen's self-organizing feature maps in the area of software reuse. The map is used for structuring the contents of a library containing reusable software according to the semantic similarity of the stored software components. Experiments with different learning functions are compared. These experiments demonstrate that lateral inhibition of output units results in earlier convergence of the map as well as in better approximation of the input data.<
>