Adaptive Incentive Allocation for Influence-Aware Proactive Recommendation

Wu, Shiqing; Bai, Quan; Kang, Byeong Ho

doi:10.1007/978-3-030-29908-8_51

Cited by 5 publications

(8 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, to the best of our knowledge, only a few studies adopted ABM to model the process of incentive allocation. Wu et al proposed the Agent-based Decision-making (ADM) model that leverages ABM to model users' behaviors in the scenario where social influence and incentive co-exist simultaneously [48]. They also proposed the IPE method to identify potential influential users in unknown social networks.…”

Section: Agent-based Modeling For Users' Behaviorsmentioning

confidence: 99%

“…However, the performance of IPE could be limited since it only considers direct influence among users when identifying influential users. Different from [48], in this paper, we consider both direct and indirect influence among users to estimate users' influential ability. Meanwhile, we will investigate how to tackle the incentive allocation problem in unknown networks based on the ADM model.…”

Section: Agent-based Modeling For Users' Behaviorsmentioning

confidence: 99%

“…In this study, we deploy the Agent-based Decision-Making (ADM) model [48] to simulate users' behaviors under the simultaneous impact of incentives and social influence. The ADM model is a decentralized model that inherits the advantages of Agent-based Modeling, i.e., each user is modeled as an autonomous agent, which maintains its personal attributes and chooses the action with the highest user utility at every time step.…”

Section: Agent-based Decision-making Modelmentioning

confidence: 99%

“…One potential candidate is the record of users' historical behaviors, as an influential relationship may exist between two users if one user's behaviors are always constant with that of the other user [52]. Nevertheless, the analysis based on the behavior record could be inaccurate, since it is common in a social network that two users perform the same behaviors even there is no connection between them [48]. In this case, it is difficult to identify the true influence relationship between each pair of users, but estimating such relationships can be possible.…”

Section: Introductionmentioning

confidence: 99%

“…The recent work [48] conducted preliminary research work on modeling users' decision-making process under the impact of incentives and social influence and proposed the Agent-based Decision-making Model (ADM). In the ADM, all users are modeled as autonomous agents, and each user would make decisions based on individual utility consisting of preferences, offered incentives, and the social influence exerted by adjacent users.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Identifying Influential Users in Unknown Social Networks for Adaptive Incentive Allocation Under Budget Restriction

Li²,

Shen³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In recent years, recommendation systems have been widely applied in many domains. These systems are impotent in affecting users to choose the behavior that the system expects. Meanwhile, providing incentives has been proven to be a more proactive way to affect users' behaviors. Due to the budget limitation, the number of users who can be incentivized is restricted. In this light, we intend to utilize social influence existing among users to enhance the effect of incentivization. Through incentivizing influential users directly, their followers or friends in the social network are possibly incentivized indirectly. However, in many real-world scenarios, the topological structure of the network is usually unknown, which makes identifying influential users difficult. To tackle the aforementioned challenges, in this paper, we propose a novel algorithm specifically for exploring influential users in unknown networks, which can estimate the influential relationships among users based on their historical behaviors and without knowing the topology of the network. Meanwhile, we design an adaptive incentive allocation approach that determines incentive values based on users' preferences and their influence ability. We evaluate the performance of the proposed approaches by conducting experiments on both synthetic and real-world datasets. The experimental results demonstrate the effectiveness of the proposed approaches.

show abstract

Section: Agent-based Modeling For Users' Behaviorsmentioning

confidence: 99%

Section: Agent-based Modeling For Users' Behaviorsmentioning

confidence: 99%

Section: Agent-based Decision-making Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations