Algorithmic Representation of Viral Media Propagation
Date: Friday , April 15, 2016
Established in 1998, IMS Noida is a premier management institute that provides management education to fresh graduates seeking world-class management education and to professionals who are already employed and want to enhance their managerial skills.
The most popular way of passing the information is the media, which is passed from person to person i.e. word of mouth. The best way to go viral is the internet and therefore, social networks have become one of the most effective channels for voicing any information, message or advertisements. The success of social blogs, Twitter, Facebook or any other, has attracted the attention of many researchers.
In IMS-Noida, a Centre for Data Science has been set up, where research in this area is being taken up. The various social networking tools are being identified for media propagation and how they are advancing in tapping the data which is further being useful in data analytics. The Centre has also developed a panel for the companies looking for data analytics. This panel comprises a pool for data of Students, Alumni, and Parents of existing students, Faculty members and Staff members. The data consists of demography and behavioral feature of the respondents. These samples of data will help the companies looking for analyzing the market for digital marketing, e-Commerce and virtual media marketing.
The common perception of viral media is about being cheap, easy and massively effective. However, recent studies have revealed that the propagation often fades quickly within only few hops from the sources, counteracting the assumption on the self-perpetuating of influence considered in literature. Some of the researchers have tried to answer the following questions given below:
1. With only limited influence propagation, is massively reaching customers via viral media still affordable?
2. How to economically spend more resources to increase the spreading speed?
3. The phenomenon of influence exerted by users of such platforms on other users, and how it propagates in the network?
4. How to make viral media more cost effective?
To address the above mentioned questions, various mathematical programming techniques are being used to numerically analyze the cost effective usage of viral media propagation. There were some studies undertaken in Stanford University to see whether an algorithm can predict which popular content can become a viral content. Algorithm is a step-by-step procedure for solving a problem or accomplishing some end, especially by a computer. A finite set of unambiguous instructions with a given set of initial conditions, can be performed in a prescribed sequence to achieve a certain goal and that has a recognizable set of end conditions. There are various algorithms like Simple recursive algorithms, Backtracking algorithms, Divide-and-conquer algorithms, Dynamic programming algorithms, Greedy algorithms, and Branch-and-bound algorithms.
Various algorithmic methods have been involved and designed to represent the propagation of viral media. Kempe et al. proposed two basic diffusion models, namely independent cascade model (IC) and linear threshold model (LT). These two models and their extensions set the foundation to almost all existing algorithms to find seeding in social networks. In both models, a social network is modeled as a directed graph G= (V, E), where the vertices of V represents individuals and edges in E represent relationships and the orientations of the edges indicate the direction of influence. The LT model focuses on the threshold behavior in influence propagation.
The Linear Threshold model applied to social network helps to understand the influence of messages on social network and others. The IC model focuses on individual (and independent) interaction and influence among friends in a social network. Cascading processes are models of network diffusion used to study phenomenon concerning the spread of new trends and innovations in social networks. Each node can be in one of two states: infected (i.e., supports an idea or a product) or uninfected. Every infected node can infect its neighbors and thus, the infection, formally called a cascade, propagates through the network.
A study on Cost-effective Viral Marketing for Time-critical Campaigns in Large-scale Social Networks found that to minimize the seeding cost, we need to provide mathematical programming to find optimal seeding for medium-size networks and propose VirAds, an efficient algorithm, to tackle the problem on large scale networks. Recently, various studies have been undertaken to study the use of algorithm in predicting the impact on viral media.