Personal Information Inference in Social Networks

In modern society, social networks play an important role for online users. However, one unignorable problem behind the booming of the services is privacy issues. This paper conducts an extensive study to infer sensitive personal information from public insensitive attributes in social networks by deploying different machine learning algorithms. The results show that several algorithms can achieve about 3 times of the baseline accuracy to classify low-correlation, high-noise dataset to infer certain sensitive attribute. Limitation for each algorithm is also discussed in the paper.