Abstract: Aiming at the task assignment issue in the mobile crowd sensing platform, a task assignment model combining task demand feature extraction algorithm with user label classification method is proposed. Firstly, the task demand feature extraction algorithm is used to extract the category keywords of sensing tasks. Then, the data set is trained by multi-linear neural network and multi-kernel learning to get the classifier, and the user’s type labels are predicted by the classifier. Finally, according to the category keywords of the tasks, combined with the spatial location information and user participation, the users who have the task category labels and meet the task requirements are selected to distribute tasks. Simulation results show that the proposed model has good feasibility in terms of task matching and task assignment efficiency.

Key words: mobile crowd sensing, feature extraction, user label, multi-linear neural network