Active Learning of Model Parameters for Influence Maximization

Cao, Tianyu; Wu, Xindong; Hu, Tony; Wang, Song

doi:10.1007/978-3-642-23780-5_28

Cited by 13 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar to the work done in [11,40,1], we focus on the selection of an ideal set of nodes for the purpose of optimizing clicks and revenue to the advertiser. We divert from approaches to the problem that utilize influence spread and the theory of submodular functions as done in several papers [26,6,27,28,30,7,8,3,4,14,20,46,47] and focus on maximizing the expected gains for the advertiser. Our formulation to the problem formally known as the IM problem is novel since in addition to adopting a decision-making perspective its main goal is to maximize the expected number or clicks or revenue to the advertiser.…”

Section: Related Workmentioning

confidence: 99%

“…Influence models for the IM problem can be described as models which capture real-world propagations or the spread of information among users within a network. In addition to the diffusion models; the Linear Threshold and Independent Cascade models defined in [22], influence models that determine node and edge probabilities have been proposed in [11,40,13,16,4,5]. For the IM-RO problem we introduce the GIM equation (2) and NIM (3) as the pertinent influence models by which probabilities are updated at the end of each stage.…”

Section: Experimental Settingsmentioning

confidence: 99%

See 1 more Smart Citation

Stochastic dynamic programming heuristics for influence maximization–revenue optimization

Lawrence

Hosein

2018

Int J Data Sci Anal

View full text Add to dashboard Cite

The well-known Influence Maximization (IM) problem has been actively studied by researchers over the past decade, with emphasis on marketing and social networks. Existing research have obtained solutions to the IM problem by obtaining the influence spread and utilizing the property of submodularity. This paper is based on a novel approach to the IM problem geared towards optimizing clicks and consequently revenue within an Online Social Network (OSN). Our approach diverts from existing approaches by adopting a novel, decision-making perspective through implementing Stochastic Dynamic Programming (SDP). Thus, we define a new problem Influence Maximization-Revenue Optimization (IM-RO) and propose SDP as a method in which this problem can be solved. The SDP method has lucrative gains for an advertiser in terms of optimizing clicks and generating revenue however, one drawback to the method is its associated "curse of dimensionality" particularly for problems involving a large state space. Thus, we introduce the Lawrence Degree Heuristic (LDH), Adaptive Hill-Climbing (AHC) and Multistage Particle Swarm Optimization (MPSO) heuristics as methods which are orders of magnitude faster than the SDP method whilst achieving near-optimal results. Through a comparative analysis on various synthetic and real-world networks we present the AHC and LDH as heuristics well suited to to the IM-RO problem in terms of their accuracy, running times and scalability under ideal model parameters. In this paper we present a compelling survey on the SDP method as a practical and lucrative method for spreading information and optimizing revenue within the context of OSNs. definition in [22], the IM problem has been actively studied by researchers over the past decade and is not restricted to applications in marketing only but also in healthcare, communication, education, agriculture, and epidemiology [27,30,35,44]. For the IM problem, OSN users are represented in a graph G = (V, E), where the nodes of V represent the users and the edges in E represent the relationships between users. In [22], the problem was first defined as a discrete optimization problem and the term influence of a set of nodes A, denoted by σ (A) was defined to be the expected number of active nodes at the end of a diffusion process, given that A was this initial active node set. According to the work done in [22], the IM problem therefore seeks to determine a parameter K, that is, to find a K − node set of maximum influence; where |A|≤ K.It is an open question to compute this K − node set and expected number of active nodes σ (A) by an efficient method, but very good estimates have been proposed and obtained [6,7,30].This paper provides a novel approach to the IM problem and a formal definition to the model proposed in [19]. We divert from all other existing approaches to IM and adopt a novel decision-making perspective primarily used in shortest paths and resource allocation problems [2,29,37,39]. Thus, we define a new problem, the IM-RO problem and implement SDP as the m...

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Experimental Settingsmentioning

confidence: 99%

Stochastic dynamic programming heuristics for influence maximization–revenue optimization

Lawrence

Hosein

2018

Int J Data Sci Anal

View full text Add to dashboard Cite

show abstract

“…This is an iterative method based on the interaction of multiple agents or "particles". Each agent corresponds to a different coefficient configuration, representing a coordinate in the parameter space of the problem 4 . Apart from the coordinates themselves, the agents also have a velocity.…”

Section: B Particle-swarm Optimizationmentioning

confidence: 99%

“…This result stresses the importance of the exact computation of vertex infection probabilities. This problem was proven to be #P-complete by Cao [4].…”

Section: Introductionmentioning

confidence: 99%

The inverse infection problem

Bóta¹,

Krész²,

Pluhár³

2014

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-The applications of infection models like the Linear Threshold or the Domingos-Richardson model requires a graph weighted with infection probabilities. In many real-life applications these probabilities are unknown; therefore a systematic method for the estimation of these probabilities is required. One of the methods proposed to solve this problem, the Inverse Infection Model, was originally formulated for estimating credit default in banking applications. In this paper we are going to test the capabilities of the Inverse Infection Model in a more controlled environment. We are going to use artificially created graphs to evaluate the speed and the accuracy of estimations. We are also going to examine how approximations and heuristics can be used to improve the speed of the calculations. Finally, we will experiment with the amount of a priori information available in the model and evaluate how well this method performs if only partial information is available.

show abstract

“…This result stresses the importance of the exact computation of vertex infection probabilities. This problem was proven to be #P-complete by Cao et al [6].…”

Section: Introductionmentioning

confidence: 95%

Estimation of Edge Infection Probabilities in the Inverse Infection Problem

Bóta

Krész

Pluhár

2015

Recent Advances in Computational Optimization

View full text Add to dashboard Cite

Several methods have been proposed recently to estimate the edge infection probabilities in infection or diffusion models. In this paper we will use the framework of the Generalized Cascade Model to define the Inverse Infection Problem-the problem of calculating these probabilities. We are going to show that the problem can be reduced to an optimization task and we will give a particle swarm based method as a solution. We will show, that direct estimation of the separate edge infection values is possible, although only on small graphs with a few thousand edges. To reduce the dimensionality of the task, the edge infection values can be considered as functions of known attributes on the vertices or edges of the graph, this way only the unknown coefficients of these functions have to be estimated. We are going to evaluate our method on artificially created infection scenarios. Our main points of interest are the accuracy and stability of the estimation.

show abstract

Active Learning of Model Parameters for Influence Maximization

Cited by 13 publications

References 14 publications

Stochastic dynamic programming heuristics for influence maximization–revenue optimization

Stochastic dynamic programming heuristics for influence maximization–revenue optimization

The inverse infection problem

Estimation of Edge Infection Probabilities in the Inverse Infection Problem

Contact Info

Product

Resources

About