Abstract: There are many problems in the traditional data privacy protection algorithm, such as long execution time, high concealment and poor adaptability. The proposed network data density clustering algorithm based on privacy protection, in the big data environment, the transformation function generates a random given the big data environment of network privacy data, then use the numerical transform function of network privacy data are changed, and the numerical transform as the result of random response then according to the characteristics of the network; the network node density clustering analysis, generated in accordance with any size and shape of the clusters, through the data network privacy and network privacy quantitative information loss data structure information for effective analysis of data privacy loss amount; and finally into the real network node in the successful generation of cluster, through increasing etc. technology, complete privacy protection of network information. The simulation results show that the proposed algorithm has a better privacy protection effect, and it has a good adaptability to the attack of a variety of different background knowledge.