A tool for the mayor of Ghent: querying geospatial data on Web-scale
Keywords: AI, Artificial Intelligence, Data Retrieval, Linked Data, Semantic Web, geospatial, open data, Indexing
Supervision: Pieter Colpaert
Students: max 2
Imagine you’re the mayor of Ghent and you want a simple overview of the accessiblity of the roads: how long would it take you to create such an overview? Or maybe an overview of whether all Ghentians have an equal access to public transport and where to next plan infrastructure works? Or maybe you want to get an overview whether there are less or more shops within the city centre after the introduction of the circulation plan?
To effectively allow machines to discover data about certain locations, we need to be able to efficiently query data using certain geospatial ranges. In this thesis, you have to design a solution that allows data to be looked up on the Web given certain geospatial ranges, this should be combinable with various other data sources on the Web to allow more complex queries to be answered. This will make it possible to ask questions like “Which bars are present that serve beer in a 5 km radius around my current locations?”, or “What is the closest bike repair shop between my home and the train station?”. We will build on existing work for Web querying, such as Linked Data Fragments.
Building a solution allowing data to be queried using certain geospatial ranges. Combining this data with other fragments on data on the Web to enable more complex queries. Evaluation of this solution compared to alternative approaches. One of these approaches is the overpass turbo API This API is able to query OpenStreetMap. We want to build a more efficient solution, by making sure the client also does a part of the query-processing.