Online social systems have become an important part of everyday life. While initial examples were used to share personal content with friends more and more online social systems are also used to do business unfortunately, privacy concerns raised in the recommendation process impede the expansion of OSN users’ friend circle. Some OSN users refuse to disclose their identities and their friends’ information to the public domain. Indeed, today OSNs provide very little support to prevent unwanted messages on user walls. In order to avoid this problem in the proposed system implemented the dynamic blocking algorithm, it’s automatically block the unwanted text post message and in the classification methods using the supervised methods are implement the decision tree, Using the decision tree, its creates the more number of category, under this category its stored more data.