Spatial Analysis of Commuting Mode Choice in Guangzhou, China

Metropolitan cities in China are commonly confronted with unresolved traffic congestion issues, primarily due to rapidly increasing traffic demand. Group disparity between commuting mode choice and its spatial distribution on road networks has enabled us to examine the factors that give rise to the discrepancies and the fundamental spatial causes of traffic congestion. In recent years, mi- cro-perspective, individual, and behavior-based spatial analysis have mushroomed and been facilitated with effective tools such as tem- poral geographic information systems （T-GIS）. It is difficult to study the interrelations between transport and space on the basis of commuting mode choice since the mode choice data are invisible in a specific space such as a particular road network. Therefore, in the field of transport, the classical origin destination （OD） four-stage model （FSM） is usually employed to calculate data when studying commuting mode choice. Based on the relative principles of T-GIS and the platform of ArcGIS, this paper considers Guangzhou as a case study and develops a spatio-temporal tool to examine the daily activities of residents. Meanwhile, the traffic volume distribution in rush hours, which was analyzed according to commuting modes and how they were reflected in the road network, was scrutinized with data extracted from travel diaries. Moreover, efforts were made to explain the relationship between traffic demand and urban spatial structure. Based on the investigation, this research indicates that traffic volumes in divergent groups and on the road networks is driven by： l） the socio-economie characteristics of travelers; 2） a jobs-housing imbalance under suburbanization; 3） differences in the spatial supply of transport modes; 4） the remains of the Danwei （work unit） system and market development in China; and 5） the transition of urban spatial structure and other factors.

Coupling hydrodynamic models with GIS for storm surge simulation： application to the Yangtze Estuary and the Hangzhou Bay, China

Storm surge is one of the most serious oceanic disasters. Accurate and timely numerical prediction is one of the primary measures for disaster control. Traditional storm surge models lack of accuracy and time effects. To overcome the disadvantages, in this paper, an analytical cyclone model was first added into the Finite-Volume Coastal Ocean Model （FVCOM） consisting of high resolution, flooding and drying capabilities for 3D storm surge modeling. Then, we integrated MarineTools Pro into a geographic information system （GIS） to supplement the storm surge model. This provided end users with a friendly modeling platform and easy access to geographically referenced data that was required for the model input and output. A temporal GIS tracking analysis module was developed to create a visual path from storm surge numerical results. It was able to track the movement of a storm in space and time. MarineTools Pro＇ capabilities could assist the comprehensive understanding of complex storm events in data visualization, spatial query, and analysis of simulative results in an objective and accurate manner. The tools developed in this study further supported the idea that the coupled system could enhance productivity by providing an efficient operating environ- ment for accurate inversion or storm surge prediction. Finally, this coupled system was used to reconstruct the storm surge generated by Typhoon Agnes （No. 8114） and simulated typhoon induced-wind field and water elevations of Yangtze Estuary and Hangzhou Bay. The simulated results show good correlation with actual surveyed data. The simple operating interface of the coupled system is very convenient for users, who want to learn the usage of the storm surge model, especially for first-time users, which can save their modeling time greatly.

Joint model of user check-in activities for point-of-interest recommendation

With the rapid development of location-based networks, point-of-interest（POI） recommendation has become an important means to help people discover interesting and attractive locations, especially when users travel out of town. However, because users only check-in interaction is highly sparse, which creates a big challenge for POI recommendation. To tackle this challenge, we propose a joint probabilistic generative model called geographical temporal social content popularity（GTSCP） to imitate user check-in activities in a process of decision making, which effectively integrates the geographical influence, temporal effect, social correlation, content information and popularity impact factors to overcome the data sparsity, especially for out-of-town users. Our proposed the GTSCP supports two recommendation scenarios in a joint model, i.e., home-town recommendation and out-of-town recommendation. Experimental results show that GTSCP achieves significantly superior recommendation quality compared to other state-of-the-art POI recommendation techniques.