In semantic network component architecture is useful because of its internal availability for information integration. In this paper we represent a new approach called Multi- Factor based Classification (MFC) embedded with the traditional Iterative Classification Algorithm which takes different factors (Node information level, Range coordinate from the communication node, SI, Implicit factor of data to reach communication node) into consideration to form components. Then the result of the components is estimated based on the Equal of the node distribution, Node range per component, Web component range and required information level of each centre. Our result shows that by changing Multi-Factors we can generate components with more equally distributed nodes, change component range and Choose low information using centre.