Research & Application Of Clustering Algorithm

Clustering is an important mining method and mining task of data mining. It finds the comparability among the data from the database, and classifies the data to make different data in different kind as much as possible, and the data in the same class are as similar as possible, i.e." birds of a feather flock together ", thus optimize the implicit useful information or knowledge in the inquiry of the extensive database and discovery data, there is extensive application in a lot of fields in the data clustering.This paper has done deep studying to the existing clustering algorithm, carried on comparative analysis to the pluses and minuses of the existing clustering algorithm at the same time. On the basis of analyzing and researching, have proposed a kind of new neural network of clustering, i.e. one based on that group is optimized in the particle clustering neural network, neural network this belong to one have tutor learn the neural network, can follow clustering's sample to study independently, does not need users to input the parameter. In the algorithm design, adopt and utilize the primitive model to approach clustering's model of each space limitlessly, thus made better clustering's result. The algorithm has adopted the way of running side by side in the structural design of the neural network, after combining the particle group and optimizing algorithms and dividing a thought, have improved clustering's efficiency greatly. Have designed a new kind of neural networks of clustering. It is simple and easy to use, have faster operation tempo and higher clustering's precision, it is a kind of common neural network of clustering. Design and realize this new neural network of clustering in Java procedure, pass to the certain place of area electric power system carry the load node to gather an experiment enunciation, this paper put forward of this kind of gather a calculate way to have the higher accuracy lately. The clustering algorithm which is designed in this paper innovates certainly on the structure of clustering's neural network after combining the intelligent algorithm of clustering, have put forward a method for design and improvement of clustering neural network in the future, the neural network of clustering designed and realized can be regarded as a common tool of the number value clustering.