Recruiting the K-most influential prospective workers for crowdsourcing platforms
- Publication Type:
- Journal Article
- Citation:
- Service Oriented Computing and Applications, 2018, 12 (3-4), pp. 247 - 257
- Issue Date:
- 2018-12-01
Closed Access
| Filename | Description | Size | |||
|---|---|---|---|---|---|
| Shahsavari2018_Article_RecruitingTheK-mostInfluential.pdf | Published Version | 888.93 kB |
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
© 2018, Springer-Verlag London Ltd., part of Springer Nature. Viral marketing is widely used by businesses to achieve their marketing objectives using social media. In this work, we propose a customized crowdsourcing approach for viral marketing which aims at efficient marketing based on information propagation through a social network. We term this approach the social community-based crowdsourcing platform and integrate it with an information diffusion model to find the most efficient crowd workers. We propose an intelligent viral marketing framework (IVMF) comprising two modules to achieve this end. The first module identifies the K-most influential users in a given social network for the platform using a novel linear threshold diffusion model. The proposed model considers the different propagation behaviors of the network users in relation to different contexts. Being able to consider multiple topics in the information propagation model as opposed to only one topic makes our model more applicable to a diverse population base. Additionally, the proposed content-based improved greedy (CBIG) algorithm enhances the basic greedy algorithm by decreasing the total amount of computations required in the greedy algorithm (the total influence propagation of a unique node in any step of the greedy algorithm). The highest computational cost of the basic greedy algorithm is incurred on computing the total influence propagation of each node. The results of the experiments reveal that the number of iterations in our CBIG algorithm is much less than the basic greedy algorithm, while the precision in choosing the K influential nodes in a social network is close to the greedy algorithm. The second module of the IVMF framework, the multi-objective integer optimization model, is used to determine which social network should be targeted for viral marketing, taking into account the marketing budget. The overall IVMF framework can be used to select a social network and recruit the K-most influential crowd workers. In this paper, IVMF is exemplified in the domain of personal care industry to show its importance through a real-life case.
Please use this identifier to cite or link to this item: