A community detection method based on ranking and heat diffusion algorithm in social networks
Author(s):
Abstract:
The past decade has witnessed a rapid development of social networks.Community detection is a very important problem in social network analysis. Finding a community in a social network is to identify a set of nodes such that they interact with each other more frequently than with the nodes outside the group. Classical clustering approach, K-means, has been shown to be very efficient to detection community. However it is sensitive to the initial seeds.To solve this problem, in this study, we first find K seeds using PageRank and second extract community structure of the network using heat diffusion similarity and K-means. Using heat diffusion similarity, we can get the networks global information about any pair of vertices. The empirical study on real networks show that our algorithm in most social networks studied is better than algorithms including K-rank, K-means, BGLL, OSLOM ,LPA , Infomap and is efficient to be used for sparse large networks.
Keywords:
Language:
Persian
Published:
Information Technology on Engineering Design, Volume:8 Issue: 1, 2015
Pages:
61 to 75
https://www.magiran.com/p1712804
سامانه نویسندگان
مقالات دیگری از این نویسنده (گان)
-
ارایه یک مدل پیش بینی غیرخطی با حداکثر حاشیه با کمک اسناد توصیفی برای بهبود عملکرد سیستم های توصیه گر
مهدی رواخواه، *، یحیی فرقانی، رضا شیبانی
مجله فناوری اطلاعات در طراحی مهندسی، پاییز و زمستان 1400 -
ارائه یک الگوریتم هیبریدی از جستجوی کلاغ بهینه شده با سیستم فازی و الگوریتم جستجوی گرانشی و به کارگیری آن در آموزش شبکه عصبی رو به جلو
آزاده آروین مهر، مهدی یعقوبی*،
مجله فناوری اطلاعات در طراحی مهندسی، بهار و تابستان 1399