Article Preview
Top1. Introduction
The proliferation of internet-enabled devices such as smartphones has led to the increased usage of Online Social Networks (OSNs) for real-time information sharing (Leskovec, Backstrom, & Kleinberg, 2009; Guille, Hacid, Favre, & Zighed, 2013). This kind of information sharing helps the society in dissemination of the useful information on a large scale in a shorter duration (Bakshy, Rosenn, Marlow, & Adamic, 2012). Also, OSNs are helping for the growth of organizational businesses by finding new customer bases/marketing medium (Pham, Tran, Thipwong, & Huang, 2019) and OSNs serve as organization’s crucial decision propagation platform during disastrous events (Ngamassi, Ramakrishnan, & Rahman, 2016; Subramaniyaswamy, et al., 2017). However, along with useful information propagation and increasing the business prospects, OSNs also serve as fertile land for false information or rumor propagation on an unprecedented scale (Wen, et al., 2015). For example, in 2013, there was a rumor initiated at OSNs related to Barack Obama's injury in an explosion at the White House. This rumor has made a major crackdown on the U.S stock market amounted to U.S dollar 136.5 billion within three minutes of propagation (Domm, 2013) (Foster, 2013). This shows that the rumor spreads faster than normal information in online mediums like OSNs (Doerr, Fouz, & Friedrich, 2011). Such an exacerbated propagation causes irreversible damage to society during emergency events as a negative effect. Consequently, researches on identifying and controlling the rumors have been a rising recent interest among industry experts and academicians.
Rumor in OSNs can be defined as an information/story that is unverified or authenticity source is unknown during its circulation in the network (DiFonzo & Bordia, 2007). There have been various research works to protect the OSNs from rumors through different methodologies such as: blocking rumor spread through node blocking (Hu, Pan, Hou, & He, 2018) and link blocking (Kimura, Saito, & Motoda, 2009), defeating rumor spread through ‘anti-rumor’ information as a protective mechanism (Li, Zhu, Li, Kim, & Huang, 2013; Afassinou, 2014; Tong, et al., 2017). In a real-world situation, blocking the individuals has privacy and user agreement issues in large scale networks like OSNs (Ahn, Shehab, & Squicciarini, 2011; Huber, Weippl, Kitzler, & Goluch, 2011). So, the protective mechanism through anti-rumor information is a widely accepted and more focused solution domain for rumor containment problems (Tripathy, Bagchi, & Mehta, 2010).
When a rumor spreads in OSNs, the authorities or individuals in the network identify true information against the rumor and propagate it in the network (Ji, Liu, & Xiang, 2014). This act of defending against rumors through anti-rumor propagation protects the OSNs by breaking the rumor in the network. This defensive mechanism to protect OSNs can be studied from the defensive mechanism of social insects to protect against the pathogen in the real-world. The defensive mechanism of both possesses the same behavior such as one-to-one contact, fast-spreading of epidemics in the system of social insects (Naug & Camazine, 2002) and OSNs (Doerr, Fouz, & Friedrich, 2011), and defending protection using the set of individuals against the epidemics in social insects (Myles, 2002) as well as OSNs (Li, Zhu, Li, Kim, & Huang, 2013). Hence, the defending protection mechanism of social insects is employed in the proposed approach to control the rumors in OSNs.