中国人民大学公共管理学院讲师/助理教授,剑桥大学土地经济系博士,剑桥大学跨学科空间分析实验室成员,曾任剑桥大学土地经济系副研究员,负责欧盟 Horizon 2020 资助的城市健康项目 Emotional Cities 高级空间分析。研究特长包括土地利用感知和空间动态模拟、时空行为数据挖掘以及地理可视化,具有在中国、英国和欧盟地区的应用经验,特别关注如何通过结合机器学习、人工智能和大数据来更好地支持空间规划、治理政策制定和智能管理。
Daily trends of reopen discussion in London under Covid-19 Pandemic
Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software. A neural network embedding model is employed in delineating urban functional use from POI (Points of Interest) data. Doc2Vec model directly trains vector representations for spatial areas while considering the spatial distribution of POIs. This paper explores the functional similarity among 574 POI classes and 4836 LSOAs (Lower Layer Super Output Areas) in Greater London. Doc2Vec model outperforms other semantic models (Word2Vec, LDA and TF-IDF) in urban functional areas identification. Similarity of POI classes trained by Doc2Vec model Create your slides in Markdown - click the Slides button to check out the example. Add the publication’s full text or supplementary notes here. You can use rich formatting such as including code, math, and images.
eMOTIONAL Cities - Mapping the cities through the senses of those who make them
该项目使用神经词嵌入技术来探索城市兴趣点(POI)数据,尤其是在POI类之间的关系中。 Doc2Vec模型是Google的Quoc V. Le and Tomas Mikolov 开发的一种常见的神经词嵌入模型。该两层神经网络的输入数据是基于大伦敦40万地理分布的POI序列。可以在UKGISR 2020 论文中找到构建POI序列的具体方法。 Doc2Vec模型返回具有固定长度的574个POI类的固定长度矢量(20维)。可以通过向量之间的余弦距离来计算所有POI类对之间的相似度矩阵。为了说明高维矩阵,我们使用TensorFlow嵌入式投影仪进行可视化。您可以通过TensorFlow访问可视化界面,也可以通过下面的嵌入式小部件进行探索。
This project focuses on the crowdsourced data harvesting and data- mining of the multi-dimensional mechanisms of urban segregation combining the geo-coding of information with the rich attributes of this type of data. This project will conduct pilots at Cambridge in the UK and then compare it with prior study of Ningbo in China from an international perspective.