Cross-validation method and system for task assignment and user recruitment model for crowd-sensing

Abstract

The invention discloses a method and system for cross-validation of task allocation and user recruitment models for crowd intelligence perception. The method includes inputting multiple space-time data obtained by dividing a crowd intelligence perception task model according to region and time by using the task allocation model to be cross-validated. Subtasks, and the parameters of the agent agent determined by the user recruitment model; import the spatiotemporal subtasks into the artificial social platform where the user agent model, artificial social environment model and agent interaction model are pre-established, and start the simulation calculation to generate experimental samples ; According to the data statistics of the specified evaluation index performed by the generated experimental sample, the experimental result including the specified evaluation index is obtained. The invention can realize the cross-validation of the task allocation model and the user recruitment model, and has the advantages of strong parallel computing ability, good user customization ability, simple and efficient development process, strong practicability and wide application prospect.

Description

technical field [0001] The invention relates to a computing technology for urban crowd intelligence perception, in particular to a method and system for cross-validation of task allocation and user recruitment models for crowd intelligence perception. Background technique [0002] In crowdsensing applications, activity initiators can publish some spatial tasks in the real world, invite users with mobile smart devices to move to the target location specified by the task to perform the task, and request data related to a specific location. The mobility of users provides unprecedented opportunities for data collection and transmission through mobile devices, such as: air quality monitoring tasks initiated by urban environmental management departments, the participants in the response need to move to the designated location of the task, and use relevant sensors to collect air quality data And upload the data to the central server via the smartphone. The life cycle of crowdsensi...

AI Technical Summary

Problems solved by technology

These studies have broken some of the original stereotyped assumptions to a certain extent, but when verifying the proposed crowd-sensing user recruitment and task assignment model, it is still only possible to carry out simulation experiments on real data sets based on the inductive optimization model, and there is no real Taking into account the impact of diversity among users, temporary withdrawal of users, and unqualified data quality

Due to the huge cost of conducting large-scale real experiments, and it is a repetitive work to conduct real experiments on all newly proposed crowd sensing user recruitment and task assignment models

Images