Alon Sela

Active Viral Marketing: Incorporating Continuous Active Seeding Efforts into the Diffusion Model

Abstract Existing viral-marketing network models commonly assume a preliminary phase in which a marketer actively infects a subset of social network’s users, represented by nodes, followed by a passive viral process, in which nodes infect other nodes without external intervention. However, in real-world commercial scenarios, substantial efforts are often invested by companies to promote their products, suggesting that the adoption of products is rarely the consequence of a viral spread alone. Under this observation, this paper proposes a new diffusion model, named Active Viral Marketing, which better fits real-world marketing scenarios, where adoption of products relies on continuous active promotion efforts by the marketer. In the proposed model, the success of a marketing attempt to infect a potential customer (uninfected node), depends on the number of adopting friends (infected neighbors) of this user, assuming a user is more likely to adopt a product if more of his/her friends have already adopted it, while taking into account that social influence diminishes over time due to a memory-loss effect. The paper further proposes a set of heuristics to schedule the marketing attempts. The main idea behind these heuristics is to utilize the information on the dynamic adoption-states of neighbor nodes, in addition to the static social network topology, when choosing the next node to seed. An extensive experimentation demonstrates how the proposed seeding heuristics improve the adoption rate of products by 30%–75% in comparison to existing state-of-the-art methods that mainly rely on the network topology.

Scheduled seeding for latent viral marketing

One highly studied topic in the field of social networks is the search for influential nodes, that when seeded (i.e. infected intentionally), may infect a large portion of the network through a viral process. However, when it comes to the spread of new products, such viral processes are rather rare. Social influence is indeed an important factor when it comes to the act of adopting a new product. However, this influence is usually latent and does not trigger the purchase action by itself, it therefore requires an additional sales effort. We propose a model and a method that better fit the product adoption scenario. Our method allocates the seeding efforts not only to precise nodes but also at precise points in time, such that the product adoption rate increases. By conducting a set of empirical simulations, we show that under realistic assumptions, our method improves the product adoption rate by 25%-50%.

Comparing the diversity of information by word-of-mouth vs. web spread

Many studies have explored spreading and diffusion through complex networks. The following study examines a specific case of spreading of opinions in modern society through two spreading schemes, defined as being either through word-of-mouth (WOM), or through online search engines (WEB). We apply both modelling and real experimental results and compare the opinions people adopt through an exposure to their friend`s opinions, as opposed to the opinions they adopt when using a search engine based on the PageRank algorithm. A simulated study shows that when members in a population adopt decisions through the use of the WEB scheme, the population ends up with a few dominant views, while other views are barely expressed. In contrast, when members adopt decisions based on the WOM scheme, there is a far more diverse distribution of opinions in that population. The simulative results are further supported by an online experiment which finds that people searching information through a search engine end up with far more homogenous opinions as compared to those asking their friends.