Title :
Maximum coverage and maximum connected covering in social networks with partial topology information
Author :
Reyes, P. ; Silva, Alonso
Author_Institution :
Dept. Stat., Univ. Carlos III de Madrid (UC3M), Getafe, Spain
fDate :
April 27 2014-May 2 2014
Abstract :
Viral marketing campaigns seek to recruit the most influential individuals to cover the largest target audience. This can be modeled as the well-studied maximum coverage problem. Another related problem, called the maximum connected cover, is when recruited nodes have to be connected. This problem ensures a strong coordination among the influential nodes which are the backbone of the marketing campaign. In this work, we are interested on both of these problems. Most of the related literature assumes knowledge about the topology of the network. Even in that case, the problem is known to be NP-hard. We analyse heuristics to these models assuming different knowledge levels about the topology of the network. We quantify the difference between these heuristics and the local and global greedy algorithms.
Keywords :
computational complexity; greedy algorithms; marketing; social networking (online); NP-hard problem; global greedy algorithm; influential nodes; knowledge levels; local greedy algorithm; maximum connected covering; maximum coverage; network topology; partial topology information; social networks; viral marketing campaigns; Communication networks; Conferences; Greedy algorithms; Knowledge engineering; Network topology; Probability distribution; Topology;
Conference_Titel :
Computer Communications Workshops (INFOCOM WKSHPS), 2014 IEEE Conference on
Conference_Location :
Toronto, ON
DOI :
10.1109/INFCOMW.2014.6849337