基于HM与LWR算法的电子设备MCS推荐优化
MCS Recommendation Optimization for Electronic Equipment Based on HM and LWR Algorithms
杨玲玲1
作者信息
- 1. 河南工业贸易职业学院 信息工程学院,河南 郑州 450064
- 折叠
摘要
为了提高移动群智感知(Mobile Crowd Sensing,MCS)中数据质量,设计了一种基于混合模型(Hybrid Model,HM)与列表级排序(List-Wise Ranking,LWR)相结合的推荐方法HM-LWR.研究结果表明:确定最优参数指标为学习速率μ为 0.01,迭代 100 次,α 取值 0.5.采用HM-LWR算法模型能够较精确预测得到参与者的任务偏好情况,分配MCS任务时可以有效提升准确性与运算效率.该研究有助于提高电子设备移动群感知能力,在智慧城市领域具有很好的推广价值.
Abstract
In order to improve the data quality in Mobile Crowd Sensing(MCS),a recommendation method HM-LWR based on Hybrid Model(HM)and List-Wise Ranking(LWR)is designed.The results show that the optimal parameters are determined as learning rate μ is 0.01,iteration 100 times,α value 0.5.The HM-LWR al-gorithm model can accurately predict the task preferences of participants,and can effectively improve the accuracy and operation efficiency when assigning MCS tasks.This research is helpful to improve the mobile group perception ability of electronic devices,and has good promotion value in the field of smart city.
关键词
移动群智感知/任务推荐/协同排序/混合模型/参与者意愿Key words
mobile crowd sensing/task recommendation/collaborative sorting/hybrid model/participant in-tention引用本文复制引用
出版年
2024
NETL
NSTL
万方数据