Crowd-Based Cooperative Task Allocation via Multicriteria Optimization and Decision-Making

Abstract

As a new computing paradigm, crowd-based cooperative computing aims at effective management and the coordinated use of crowd resources. In crowd-based cooperative task allocation (CBCTA), it is necessary to ensure the suitability and high-quality collaboration of resources for computer supported cooperative work. Generally, the high matching rate between resource and task requirements can achieve the optimal parameter configuration, whereas high-quality collaboration ensures the quality and success rates of crowd-based cooperative task. This article proposes a methodology to optimize the resource allocation model for solving CBCTA problems in a cost-efficient, requirements adapted fashion. Specifically, the proposed methodology hinges on evolutionary heuristics to find proper resources that optimally balance matching rate and collaborative quality. We also present suitable metrics to quantify the aforementioned targets. Furthermore, the obtained solutions are ranked based on multicriteria decision making to provide a flexible design choice for decision-makers. Different scales of CBCTA problems are conducted to illustrate the value of the proposed methodology. The experimental results show that the proposed methodology is effective and feasible.