Crowdsourcing incentive mechanisms for cross-platform tasks: A weighted average maximization approach

Crowdsourcing refers to the practice of outsourcing tasks previously performed by internal employees of an enterprise or organization to the general public through the internet in a free and voluntary manner, resulting in a mutually beneficial outcome. The incentive mechanism is a critical aspect of crowdsourcing computing. However, existing research mainly focuses on the incentive mechanism of crowdsourcing tasks on a single platform, while crowdsourcing tasks have complex and cross-domain attributes. In actual task requests, multiple crowdsourcing platforms participate in task execution due to geographic, capability, and task attribute constraints, each with its own unique characteristics and attributes. To address the collaboration problem between different platforms in crowdsourcing task allocation, we propose a Multi -Unit and MultiPlatform (MUMP) incentive mechanism based on task interactions, where we first model the problem, design an optimization goal of maximizing the weighted average for cross-platform crowdsourcing tasks, and then propose a feasible budget algorithm with platform weight based on greedy ordering, which achieves an approximation rate. Finally, experimental results demonstrate that the proposed incentive mechanism algorithm outperforms the latest algorithm.