iVizTRANS: Interactive visual learning for home and work place detection from massive public transportation data
- Resource Type
- Conference
- Authors
- Yu, Liang; Wu, Wei; Li, Xiaohui; Li, Guangxia; Ng, Wee Siong; Ng, See-Kiong; Huang, Zhongwen; Arunan, Anushiya; Watt, Hui Min
- Source
- 2015 IEEE Conference on Visual Analytics Science and Technology (VAST) Visual Analytics Science and Technology (VAST), 2015 IEEE Conference on. :49-56 Oct, 2015
- Subject
- Computing and Processing
Data visualization
Smart cards
Feature extraction
Cities and towns
Spatiotemporal phenomena
Visualization
Clustering algorithms
Smart card data
origin-destination (OD)
spatiotemporal visualization
clustering
machine learning
- Language
Using transport smart card transaction data to understand the homework dynamics of a city for urban planning is emerging as an alternative to traditional surveys which may be conducted every few years are no longer effective and efficient for the rapidly transforming modern cities. As commuters travel patterns are highly diverse, existing rule-based methods are not fully adequate. In this paper, we present iVizTRANS - a tool which combines an interactive visual analytics (VA) component to aid urban planners to analyse complex travel patterns and decipher activity locations for single public transport commuters. It is coupled with a machine learning component that iteratively learns from the planners classifications to train a classifier. The classifier is then applied to the city-wide smart card data to derive the dynamics for all public transport commuters. Our evaluation shows it outperforms the rule-based methods in previous work.