Finding Similar Users in Social Networks by Using the Neural-Based Skyline Region

Abstract

Abstract:Finding similar users in a community network is a very important part of the recommendation system. Traditionally, algorithms of this kind are based on a single condition to search for similar users. In recent years, some scholars have proposed the popular multi-criteria algorithm Skyline Query to search for similar users. However, their proposed methods might have found users not similar to the target user and are subject to the problem of slow execution. To solve these issues, this paper introduces the concepts of neural-based skyline region and neural network. The simulation results have verified the effectiveness and execution efficiency of the target method.

Finding m-similar users in social networks using the m-representative skyline query

“…Most conventional recommendation systems use cosine similarity (Lu et al , 2012; Shamir and Tishby, 2010) or k -means algorithms (Li et al , 2008) to combine the multi-dimensional data of a user into a single score, and then search for similar users based on this score. However, this approach is actually highly illogical – as the dimensions of user data are independent and unrelated, they should not be combined into a single indicator (Hou et al , 2015). For example, Tables I and II are largely similar, with the exception of Table II having an additional User G. Both Users G and A have similar ratings for the Tokyo National Museum but different ratings for the other two Tokyo attractions.…”

“…The second method, developed by Hou et al (2015), uses the skyline region concept to identify users similar to A. The first step is to identify the skyline users of A, and then establish a threshold σ to map out a skyline region, as shown on the dotted line of the Figure 3.…”

“…In this work, we adopt the concept of depth- k (Chen and Lee, 2011a; Chiu et al , 2015; Hou et al , 2015) to assist in the selection of m -representative skyline results. We begin by using the depth- k skyline query to determine a depth score for each user in the selection of the m -users with the highest degree of similarity to the target user.…”