Facebook, Twitter, LinkedIn, Instagram are well known social networking sites. These sites gather data regarding their user’s interests, disinterests, location, profession etc. This data may contain sensitive information about the user. This data is used for research and business purposes. So this social network data should be made anonymize before it is made available to the third parties. This paper deals with community preservation in dynamic social networks and for this it considers social network at two different time periods. Fast k-degree anonymity algorithm is used to anonymize the initial social network and to produce the anonymized social network. To form communities from the initial and anonymized social networks Louvain community detection method is used. Then the community preservation is performed between communities from initial and anonymized graphs. And the percentage community preservation is obtained for different time periods.