基于多Agent的众包任务推荐系统建模与仿真

发布时间：2018-07-04 11:48

本文选题：众包设计 + 任务推荐　；参考：《计算机工程与科学》2017年05期

【摘要】：为了让众包平台用户更方便准确地搜寻到合适任务,促进其能力水平提升,解决众包任务推荐动态性等问题,提出了一种基于多Agent的众包任务推荐系统。首先,基于众包平台建立多Agent任务推荐模型,提出了模型设计思路与模型框架,并进一步阐述了各Agent功能、相互作用关系与相关算法;其次,提出众包用户能力水平提升相关算法;最后,利用NetLogo仿真软件进行验证。结果表明,众包任务推荐系统可对用户能力水平的提升起到促进作用,证明了在众包平台引入推荐系统的必要性。并且分析了多Agent技术可提升推荐系统的动态性、智能性与灵活性等整体性能,促进了众包平台数据的管理与维护。
[Abstract]:In order to make the users of crowdsourcing platform more convenient and accurate to find suitable tasks, to promote their ability to improve, and to solve the dynamic problems of crowdsourcing task recommendation, a crowdsourcing task recommendation system based on multiple Agent is proposed. First of all, based on the crowdsourcing platform, the multi-task recommendation model of Agent is established, the model design idea and model framework are proposed, and the functions of each Agent, interaction relations and related algorithms are further elaborated. The related algorithms of crowdsourcing user capability enhancement are proposed. Finally, NetLogo simulation software is used to verify this algorithm. The results show that the crowdsourcing task recommendation system can promote the improvement of the user's ability, which proves the necessity of introducing the recommendation system in the crowdsourcing platform. It also analyzes that multi-Agent technology can enhance the dynamic, intelligent and flexible performance of recommendation system, and promote the data management and maintenance of crowdsourcing platform.
【作者单位】：天津大学机械工程学院;
【基金】：国家863计划(2015BAF32B03,2013AA040605)
【分类号】：TP391.3

【相似文献】