加入收藏夹
联系我们
关于本站
个人主页
西电导航
西安电子科技大学
    当前位置:首页>>学术讲座
机器学习应用在城市规划和智慧城市方面的研究
时间:2017-11-14 11:59    点击:   所属单位:空间科学与技术学院
讲座名称 机器学习应用在城市规划和智慧城市方面的研究
讲座时间 2017-11-18 16:30:00
讲座地点 南校区G240报告厅
讲座人 Dr. Hu Nan
讲座人介绍
Dr. Hu Nan is a scientist at the computing science department of Institute of High Performance Computing, A*STAR, Singapore. He obtained his B. Eng degree with honor and Ph.D both from School of Computer Engineering, NTU. His main research expertise is on agent-based modelling and simulation, particularly for crowd and other complex systems involving human decision and cognitive process. He was one of the lead researchers in projects covering various topics such as crowd simulation, automatic optimization of agent-based model and video analytics of crowd features. He also developed capabilities in applied machine-learning techniques particularly for urban planning and published papers in top-tier journals such as Plos One and Land Use Policy.
讲座内容 We built a model capable of accounting for the impact of land-use features at two differing levels of granularity on transport demand – land-use sector (more general) and amenity structure (more granular).  Results from the model reveal that amenity-related features are better predictors than general land-use information.  More intriguingly, the model corroborates that the development of "regional centres" can reduce demand for transport.  Further, for the regional centre conceptual plan for 2030, the model yields two quantitative critical points: (a) public transport demand starts to fall after initial increase when the total amenities density (particularly deployed in the regional centres) increases by 55%, and (b) public transport demand will fall below current levels should amenities density increases by 110% and beyond.

The model put forward in this work can assist planners to better strategize land-use configurations, in particular the amenity resource distribution, to influence and shape public transportation demand.
转载请注明出处:西安电子科技大学学术信息网
如果您有学术信息或学术动态,欢迎投稿。我们将在第一时间确认并收录,投稿邮箱: meeting@xidian.edu.cn
Copyright © 2011-2017 西安电子科技大学 
开发维护:电子工程学院网络信息中心  管理员:meeting@xidian.edu.cn 站长统计: