Finding similar users in social networks

Abstract

We consider a system where users wish to find similar users. To model similarity, we assume the existence of a set of queries, and two users are deemed similar if their answers to these queries are (mostly) identical: each user has a vector of preferences, and two users are similar if their preference vectors differ in only a few coordinates. The preferences are unknown to the system initially, and the goal of the algorithm is to classify the users into classes of roughly the same preferences with the least possible number of queries presented to any user. We prove nearly matching lower and upper bounds on that problem. Specifically, we present an "anytime" algorithm that maintains a partition of the users, and the quality of the partition improves over time: let n be the number of users. At time T , groups ofÕ (n/T ) users with the same preferences will be separated (with high probability) if they differ in sufficiently many queries. We present a lower bound that matches the upper bound, up to a constant factor, for nearly all possible distances between user groups.

A Particle Swarm Optimization Based Approach for Finding Similar Users on FacebookA Particle Swarm Optimization Based Approach for Finding Similar Users on Facebook

“…For example, Aviv Nisgav, Boaz Patt-Shamir [6] has considered each user as a vector of preferences (answers to queries). They consider two users similar if their preference vectors differ in only a few coordinates.…”

“…Finding similar people is useful in many applications like recommender systems, predicting stable marriage or relationship, recruiting employees, etc. In their work, Aviv Nisgav, Boaz Patt-Shamir [6] has used a query based classification algorithm to find similarity between two users. On the other hand, Pasquale De Meo et al [7] have proposed an approach based on the knowledge of social ties existing among users, and the analysis of activities in which users are involved, in order to estimate the similarity of two users.…”

User Preference-oriented Collaborative Recommendation Algorithm in e-commerce

“…Scholars have done some research on using the network to get the user preference to form a network in which users' preferences is similar. Nisgav [3] etc. considered how a user find its similar ones in the social network and put forward to judge whether the two users are similar according to the answer of the question.…”

Collaborative scoring with dishonest participants

“…Finally, Nisgav and Patt-Shamir [13] have recently developed algorithms for partitioning users based on their preferences. Their goal is not to determine preferences, but instead to group users into sets with similar preferences.…”

“…To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. SPAA'10, June [13][14][15]2010, Thira, Santorini, Greece. Copyright 2010 ACM 978-1-4503-0079-7/10/06 ...$10.00.…”