Public Transit Route Planning through Lightweight Linked Data Interfaces
How cost-efficient is data publishing compared to an origin destination API? As end-users want to plan public transit journeys based on parameters that exceed a data publisher’s imagination, Linked Connections becomes provably a more cost-efficient HTTP interface when reuse increases.
Open Transport Data for maximising reuse in multimodal route planners: a study in Flanders
A paper discussing (open) data source interoperability at the Department of Transport and Public Works on 4 levels: legal, syntactic, semantic and querying. 10 data challenges we identified to which possible solutions were formulated. The effort needed to reuse existing public datasets today is high, yet we see the first evidence of datasets being reused in a legally and syntactically interoperable way.
Constraints for a large-scale ITS data-sharing system: a use case in the city of Ghent
Thanks to architectural constraints adopted by its stakeholders, the World Wide Web was able to scale up to its current size. To realize the ITS directive, which stimulates sharing data on large scale between different parties across Europe, a large-scale information system is needed as well. We discuss three constraints which lie at the basis of the success of the Web, and apply these to transport data publishing: stateless interaction, cacheability and a uniform interface. The city of Ghent implemented these constraints for publishing the dynamic capacity of the parking sites. The information system, allowing federated queries from the browser, achieves a good user perceived performance, a good network efficiency, achieves a scalable server infrastructure, and enables a simple to reuse dataset. To persist such a transport data information system, still a well-maintained Linked Data vocabulary is needed. We propose to add these URIs into the DATEX2 specification.
Predicting train occupancies based on query logs and external data sources
A paper trying to predict how crowded your next train will be using a machine learning algorithm.