Predicting and Identifying Missing Node Information in Social Networks

Abstract

In recent years, social networks have surged in popularity. One key aspect of social network research is identifying important missing information which is not explicitly represented in the network, or is not visible to all. To date, this line of research typically focused on finding the connections that are missing between nodes, a challenge typically termed as the Link Prediction Problem. This paper introduces the Missing Node Identification problem where missing members in the social network structure must be identified. In this problem, indications of missing nodes are assumed to exist. Given these indications and a partial network, we must assess which indications originate from the same missing node and determine the full network structure.Towards solving this problem, we present the MISC Algorithm (Missing node Identification by Spectral Clustering), an approach based on a spectral clustering algorithm, combined with nodes' pairwise affinity measures which were adopted from link prediction research. We evaluate the performance of our approach in different problem settings and scenarios, using real life data from Facebook. The results show that our approach has beneficial results and can be effective in solving the Missing Node Identification Problem. In addition, this paper also presents R-MISC which uses a sparse matrix representation, efficient algorithms for calculating the nodes' pairwise affinity and a proprietary dimension reduction technique, to enable scaling the MISC algorithm to large networks of more than 100,000 nodes. Last, we consider problem settings where some of the indications are unknown. Two algorithms are suggested for this problem -Speculative MISC, based on MISC, and Missing Link Completion, based on classical link prediction literature. We show that Speculative MISC outperforms Missing Link Completion.

Solving the missing node problem using structure and attribute information

“…Our work extends Eyal et al's MISC algorithm [13]. The MISC algorithm focused on a specific variation of the missing nodes problem where the missing nodes requiring identification are "friends" of known nodes.…”

“…In this problem setting, the nodes of the network are known, and unknown links are derived from existing network information, including node information. Most recently a new missing node identification problem [13] was introduced, which locates and identifies missing nodes within the network. Previous studies have shown that combining the nodes' attributes can be effective when inferring missing links or attributes [7], [14].…”

“…Recent algorithms, MISC [13] and KronEM [10], which were developed to solve similar problems, used the structure of the network, but did not consider information about specific nodes. Our work extends Eyal et al's MISC algorithm [13].…”

“…Towards this goal, we present two algorithms suitable for solving this missing node problem: SAMI-A (Structure and Attributes Missing node Identification using Attributes' similarity) and SAMI-N (Structure and Attributes Missing node Identification using social-attribute Network). The first algorithm, SAMI-A, calculates a weighted sum between two affinity components, one based on the network graph structure, as was the case in previous work [13], and a new measure based on common attributes within known nodes. The second algorithm, SAMI-N, combines the known nodes' attributes data into a Social-Attribute Network (SAN) -a data structure that was previously developed [7], [15], [16].…”

Unveiling Privacy Setting Breaches in Online Social Networks

“…There are two major research directions on the privacy and security issues in OSNs: (1) to reveal the privacy threats in OSNs by conducting surveys [16,17] and proposing attack models [26], information inference algorithms [6,8,9,13,14,19,28], de-anonymization algorithms [4,21], and re-identification algorithms [27]; and (2) to reinforce users' privacy by redesigning the OSN system structure [5,10,20,23] and conducting anonymization [22,25]. This paper investigates the privacy setting breaches, which belongs to (1).…”