Influence- and Interest-based Worker Recruitment in Crowdsourcing using Online Social Networks

TL;DR

The paper addresses the cold-start and low participation problems in Mobile Crowdsourcing by proposing IIWRS, an OSN-assisted recruitment system. It introduces a GA-based, group-aware Influence Maximization stage to select influencer groups by maximizing the group score $R(g) = \sqrt[3]{D^V(g) \times I^V(g) \times U^V(g)}$, followed by information diffusion under the Independent Cascade model to form a pool of candidate workers, and a dynamic, interest-aware recruitment stage that maximizes the per-worker QoS ${QoS}_j^W = \sqrt[4]{ {RE}_j^W \times {IL}_j^W \times {\tau}_j^W \times Rep_j^W }$ with a travel-time dependent $\tau_j^W$. The approach is evaluated on real OSN data (Twitter-derived) and MCS task datasets, demonstrating up to 15% and 20% improvements in IM metrics and up to 88x QoS gains over baselines like SWRS, GRS, and their dynamic variants. The work shows that integrating group-based influencer selection, interest alignment, diffusion-based pooling, and dynamic substitution yields robust, scalable recruitment capable of mitigating cold-start and low-task-acceptance challenges in MCS. The proposed framework has practical impact for deploying OSN-driven recruitment in location-based and crowdsourcing tasks with heterogeneous domain needs.

Abstract

Workers recruitment remains a significant issue in Mobile Crowdsourcing (MCS), where the aim is to recruit a group of workers that maximizes the expected Quality of Service (QoS). Current recruitment systems assume that a pre-defined pool of workers is available. However, this assumption is not always true, especially in cold-start situations, where a new MCS task has just been released. Additionally, studies show that up to 96\% of the available candidates are usually not willing to perform the assigned tasks. To tackle these issues, recent works use Online Social Networks (OSNs) and Influence Maximization (IM) to advertise about the desired MCS tasks through influencers, aiming to build larger pools. However, these works suffer from several limitations, such as 1) the lack of group-based selection methods when choosing influencers, 2) the lack of a well-defined worker recruitment process following IM, 3) and the non-dynamicity of the recruitment process, where the workers who refuse to perform the task are not substituted. In this paper, an Influence- and Interest-based Worker Recruitment System (IIWRS), using OSNs, is proposed. The proposed system has two main components: 1) an MCS-, group-, and interest-based IM approach, using a Genetic Algorithm, to select a set of influencers from the network to advertise about the MCS tasks, and 2) a dynamic worker recruitment process which considers the social attributes of workers, and is able to substitute those who do not accept to perform the assigned tasks. Empirical studies are performed using real-life datasets, while comparing IIWRS with existing benchmarks.

Figures (9)

• Figure 1: General overview of the proposed system (IIWRS).
• Figure 2: An overview of the architecture of the proposed system. Given a set of tasks, submitted by task publishers, the management platform selects a set of influencers that maximizes the spread of influence to a targeted group in the network (here different colors indicate different interests). Influenced users are those who register in the system, representing the pool of candidate workers from which the management platform recruits workers for each task, and collects the submitted reports by the workers. Reports are assessed for quality assurance and sent to the task publishers.
• Figure 3: Example of a simple directed network.
• Figure 4: Flowchart of the MCS-, Group-, and Interest-based IM process.
• Figure 5: Flowchart of the recruitment approach.