With the increasing dependency of all aspects of social life on Internet, the data on the internet is becoming more and more massive, and also more complex. This heterogeneous and dynamic information which is also distributed makes the traditional data mining unable to achieve actual requirements. This paper proposes an improved iterative algorithm for web data mining: combining iteration method with a parallel algorithm. And a web data mining mode is set up by the algorithm with the idea of local computing of storage nodes, which supports the parallel association rule. Experimental results show that this mode can improve the efficiency of web data mining and its implementation rate will rise as the data quantity increases.

web mining  /  iterative algorithm  /  parallel algorithmic  /  local computing