Software Crowdsourcing Task Allocation Algorithm Based on Dynamic Utility

Abstract

For crowdsourcing software development, poor matching between the workers' development skills and the requirements for the assigned tasks results in poor quality task completion. In this paper, we propose a dynamic utility task allocation algorithm (DUTA), a software crowdsourcing task allocation algorithm based on the dynamic utility. First, using the attributes provided by the worker registration information, we estimate the initial value of a worker's development abilities based on the attribute weights and levels. Second, the worker's development capabilities are calculated based on his or her history of the completed tasks, including task complexity, quality, and development efficiency. The worker's record of development capability is updated dynamically. Then, based on the skill weights, we calculate the degree to which the task requirements matche the worker's skills. Finally, we take the product of the worker's development ability and the degree of skill matching as the allocation utility, and the total utility is maximized as the optimization goal. To solve the optimal match between tasks and workers, we use the Kuhn-Munkres algorithm with a weighted bipartite graph. Our experimental results showed that DUTA provided good performance in total allocation utility, with a better task allocation success rate than the algorithm based on the user reliability. The DUTA achieved an average allocation accuracy of up to 85.63%, demonstrating that it can guide task assignments effectively.