Forward Backward Similarity Search in Knowledge Networks

Similarity search is a fundamental problem in social and knowledge networks like GitHub, DBLP, Wikipedia, etc. Existing network similarity measures are limited because they only consider similarity from the perspective of the query node. However, due to the complicated topology of real-world networks, ignoring the preferences of target nodes often results in odd or unintuitive performance. In this work, we propose a dual perspective similarity metric called Forward Backward Similarity (FBS) that efficiently computes topological similarity from the perspective of both the query node and the perspective of candidate nodes. The effectiveness of our method is evaluated by traditional quantitative ranking metrics and large-scale human judgement on four large real world networks. The proposed method matches human preference and outperforms other similarity search algorithms on community overlap and link prediction. Finally, we demonstrate top-5 rankings for five famous researchers on an academic collaboration network to illustrate how our approach captures semantics more intuitively than other approaches. Computing the similarity of two or more objects in an information network is the main focus of a large amount of scientific research and technological development. Friendship recommendation in social networks is one example, but web search, community detection, general link prediction, list augmentation, and dozens of other application areas are all singularly dependent upon some notion of similarly in the underlying networks. Similarity is multi-faceted; various traits can be used to determine similarity depending on the specific problem domain. Entire fields of research are dedicated to the development of algorithms that effectively and efficiently retrieve objects similar to some query-object, e.g., information retrieval, computer vision, and databases (broadly speaking). Researchers and practitioners understand that network topology plays a critical role in the identification of object similarity [26, 27, 35]. An appreciation of the topological features has led to the development of models of network growth, clustering, prediction, and classification. Given a query vertex u, what we need is a network similarity metric that finds a target vertex v to be similar if they satisfy the following criteria: 1. u is highly connected to v, and 2. v is highly connected to u A typical approach used to compute personalized search is to measure the similarity between some query node and a set of candidate target nodes (maybe all other nodes). After the similarities of the candidate nodes have been found, the user is typically presented with a top-K list of candidate nodes ordered by their similarity scores. For example, in citation networks Case et al. had previously defined six citation behaviors [5], which we simplify into two categories: a) intra-domain citations and b) cross-domain citations. Intra-domain references often include related prior work that is directly related to the referencing paper, and are the type of references that a reader would expect to see included in the experimental comparison section of the referencing paper. On the other hand, cross-domain citations often represent paradigms, platforms, and data sets that come from a separate, loosely-related area. For example, the closely related references of this paper include references to personal PageRank [11], SimRank [12] and personal SALSA [3]; while the loosely related references of this paper include references to DBLP [21], and ArnetMiner [48] datasets, or the reference to the Spark system [53] 1 ar X iv :1 61 1. 09 31 6v 1 [ cs .S I] 2 8 N ov 2 01 6