Ruocheng Guo

Diffusion in Social Networks

This book presents the leading models of social network diffusion that are used to demonstrate the spread of disease, ideas, and behavior. It introduces diffusion models from the fields of computer science (independent cascade and linear threshold), sociology (tipping models), physics (voter models), biology (evolutionary models), and epidemiology (SIR/SIS and related models). A variety of properties and problems related to these models are discussed including identifying seeds sets to initiate diffusion, game theoretic problems, predicting diffusion events, and more. The book explores numerous connections between social network diffusion research and artificial intelligence through topics such as agent-based modeling, logic programming, game theory, learning, and data mining. The book also surveys key empirical results in social network diffusion, and reviews the classic and cutting-edge research with a focus on open problems.

Evolutionary Graph Theory

Evolutionary graph theory (EGT), studies the ability of a mutant gene to overtake a finite structured population. In this chapter, we describe the original framework for EGT and the major work that has followed it. Here, we will study the calculation of the “fixation probability”—the probability of a mutant taking over a population and focuses on game-theoretic applications. We look at varying topics such as alternate evolutionary dynamics, time to fixation, special topological cases, and game theoretic results.

The Independent Cascade and Linear Threshold Models

In this chapter, we focus on perhaps the two most prevalent diffusion models in computer science—the independent cascade and linear threshold models. We describe different properties of these models and how these properties affect solving problems such as influence maximization and influence spread. We describe approaches to address influence maximization problem in independent cascade model and linear threshold model that rely on the maximization of submodular functions—as well as extensions to these approaches for larger datasets.

Temporal Analysis of Influence to Predict Users’ Adoption in Online Social Networks

Different measures have been proposed to predict whether individuals will adopt a new behavior in online social networks, given the influence produced by their neighbors. In this paper, we show one can achieve significant improvement over these standard measures, extending them to consider a pair of time constraints. These constraints provide a better proxy for social influence, showing a stronger correlation to the probability of influence as well as the ability to predict influence.

Understanding and forecasting lifecycle events in information cascades

Abstract Most social network sites allow users to reshare a piece of information posted by a user. As time progresses, the cascade of reshares grows, eventually saturating after a certain time period. While previous studies have focused heavily on one aspect of the cascade phenomenon, specifically predicting when the cascade would go viral, in this paper, we take a more holistic approach by analyzing the occurrence of two events within the cascade lifecycle—the period of maximum growth in terms of surge in reshares and the period where the cascade starts declining in adoption. We address the challenges in identifying these periods and then proceed to make a comparative analysis of these periods from the perspective of network topology. We study the effect of several node-centric structural measures on the reshare responses using Granger causality which helps us quantify the significance of the network measures and understand the extent to which the network topology impacts the growth dynamics. This evaluation is performed on a dataset of 7407 cascades extracted from the Weibo social network. Using our causality framework, we found that an entropy measure based on nodal degree causally affects the occurrence of these events in 93.95% of cascades. Surprisingly, this outperformed clustering coefficient and PageRank which we hypothesized would be more indicative of the growth dynamics based on earlier studies. We also extend the Granger causality Vector Autoregression model to forecast the times at which the events occur in the cascade lifecycle.

An empirical evaluation of social influence metrics

Predicting when an individual will adopt a new behavior is an important problem in application domains such as marketing and public health. This paper examines the performance of a wide variety of social network based measurements proposed in the literature - which have not been previously compared directly. We study the probability of an individual becoming influenced based on measurements derived from neighborhood (i.e. number of influencers, personal network exposure), structural diversity, locality, temporal measures, cascade measures, and metadata. We also examine the ability to predict influence based on choice of classifier and how the ratio of positive to negative samples in both training and testing affect prediction results - further enabling practical use of these concepts for social influence applications.

A comparison of methods for cascade prediction

Information cascades exist in a wide variety of platforms on Internet. A very important real-world problem is to identify which information cascades can “go viral”. A system addressing this problem can be used in a variety of applications including public health, marketing and counter-terrorism. As a cascade can be considered as compound of the social network and the time series. However, in related literature where methods for solving the cascade prediction problem were proposed, the experimental settings were often limited to only a single metric for a specific problem formulation. Moreover, little attention was paid to the run time of those methods. In this paper, we first formulate the cascade prediction problem as both classification and regression. Then we compare three categories of cascade prediction methods: centrality based, feature based and point process based. We carry out the comparison through evaluation of the methods by both accuracy metrics and run time. The results show that feature based methods can outperform others in terms of prediction accuracy but suffer from heavy overhead especially for large datasets. While point process based methods can also run into issue of long run time when the model can not well adapt to the data. This paper seeks to address issues in order to allow developers of systems for social network analysis to select the most appropriate method for predicting viral information cascades.

Examining Diffusion in the Real World

Throughout this book, we studied a variety of diffusion models that are commonly seen in the literature of computer science, physics, and biology. In this chapter, we study diffusion processes from a data-driven perspective—specifiably reviewing the early identification of information cascades that will diffuse through a large portion of the network.

The SIR Model and Identification of Spreaders

In this chapter, we review the classic susceptible-infected-recovered (SIR) model for disease spread as applied to a social network. In particular, we look at the problem of identifying nodes that are “spreaders” which cause a large part of the population to become infected under this model. To do so, we survey a variety of nodal measures based on centrality (degree, betweenness, etc.) and other methods (shell decomposition, nearest neighbor analysis, etc.). We then present a set of experiments that illustrate the relation of these nodal measures to spreading under the SIR model.

Logic Programming Based Diffusion Models

In this chapter, we first show that the well-known generalized annotated program (GAP) paradigm can be used to express many existing diffusion models that can consider not only the topology of the social network, but attributes of the nodes and edges as well. We then define a class of problems called Social Network Diffusion Optimization Problems (SNDOPs). In this chapter, we show how various diffusion processes can be embedded as GAP’s and then study the algorithmic and complexity issues associated with SDNOP’s. Experimental results are also included.