The similarity features of network users reflect the preferences of users

in order to improve the adaptive recommendation ability of network information

A similarity feature search algorithm for network users based on dynamic clustering analysis is proposed. Non-stationary random sequence analysis method is used to construct the network user information transfer model

and the association rule feature quantity which reflects the network user behavior feature is extracted

and the association information annotation method is used to retrieve the network user similarity behavior characteristic information. Based on the dynamic clustering analysis of the extracted network users similarity feature information

the fuzzy clustering method is used for adaptive classification and recognition of the similarity behavior feature information

so as to improve the ability of adaptive searching and information processing of the network users similarity features. The simulation results show that the proposed method is more accurate in searching network users similarity features

and it has a better ability of clustering and recognition of network users similarity behavior information

thus improving the ability of optimizing and recommending network information.