Comparison of algorithms for patent documents clusterization

Ever increasing number of patents makes impossible to find and analyze relevant documents manually. Various software tools have been developed in the patent field. They could analyze individual patents as well as patent portfolios; retrieve patents and make basic statistics as well as visualize, map and landscape the same data. The essential function any tool should provide is patent clustering. There have been many different clustering approaches. In this paper we compare performances of k-means, the neural-gas, fuzzy c-means and ronn clustering technique when used on patent data set that was also clustered by the experts.