Open AccessClustering Analysis of Urban Fabric Detection Based on Mobile Traffic Data
Author(s) -
Chuangfei Liu,
Chang Liu,
Fuqiang Liu,
Jianyao Hu
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1453/1/012158
Subject(s) - cluster analysis , mobile phone , computer science , ground truth , snapshot (computer storage) , unsupervised learning , k means clustering , deep learning , artificial intelligence , geography , data mining , telecommunications , database
The rapid development of city makes it complicated to analyse urban structure. It’s difficult to learn city ecosystem using traditional methods including interview and survey. A city generates lots of data every day, which could tell city’s dynamic fabric. The moving patterns of citizens could be illustrated through analysing traffic data on their mobile phones, since those who carry a mobile phone possess a large percentage of the urban population. Thus, we can get the designated urban area’s functions through the moving pattern analysis. In this paper, we explore the fabric of city using cluster analysis based on deep learning with mobile phone communication data. We get the inspiration from image processing and build communication snapshot map to represent each region. After extracting features using deep learning method, we use unsupervised learning to find similar regions of the city. The clustering analysis result is examined by the ground truth data.