Knowledge Graphs (KG) on the Semantic Web are a representation of linked data that are collected from structured and unstructured sources and used by question-answering systems, for example, Google’s search engine. Places and geo-referenced information in general are an important part of the linked data cloud, with services such as GeoNames and DBpedia providing millions of geographic entities and acting as hubs that link many other data sets. Geographic questions about places and geo-referenced information pose several challenges to these systems, from understanding the natural language query expression to translating it to meaningful queries against the KG. This project will build on ongoing work in two main areas of research: 1) extending KG embedding techniques with the geographic dimension. (Approaches to address incompleteness of KG propose the application of graph learning methods by embedding entities and relations into continuous vector spaces for link prediction and query answering.) Representation of the extent of geographic entities and spatial relations explicitly in the KG should facilitate the application of quantitative and qualitative spatial reasoning over the graph, which otherwise would not be possible, and 2) geographic query disambiguation, involving recognition and mapping of location and spatial relations expressions. This step will build on existing studies in this area but with a focus on the integration of and mediation between location expressions provided in different KGs. Previous work considered the use of gazetteers and other non-formal resources in recognising geographic entities and the location in question. However, the problem is complicated when information about the same objects or regions in space are defined in different sources, to different resolutions and with different ontologies. The project will propose methods that will employ spatial reasoning for handling possible incompleteness and vagueness in the graph, for example, by relaxation of spatial constraints in the query and comparing the similarity of location expressions in the graph. Generic and geographic KG data sets for evaluating the proposed methods are available. Evaluation experiments will also consider benchmark data sets and queries provided in the recent literature.
Keywords: Knowledge Graph, Semantic Web, Question Answering (QA), Geographic QA, Spatial Reasoning
Contact for information on the project: [Email Address Removed]
Academic criteria: A 2:1 Honours undergraduate degree or a master's degree, in computing or a related subject. Applicants with appropriate professional experience are also considered. Degree-level mathematics (or equivalent) is required for research in some project areas.
Applicants for whom English is not their first language must demonstrate proficiency by obtaining an IELTS score of at least 6.5 overall, with a minimum of 6.0 in each skills component.
How to apply:
Please contact the supervisors of the project prior to submitting your application to discuss and develop an individual research proposal that builds on the information provided in this advert. Once you have developed the proposal with support from the supervisors, please submit your application following the instructions provided below
Please submit your application before the application deadline 29th April 2022 via Computer Science and Informatics - Study - Cardiff University
In order to be considered candidates must submit the following information:
- Supporting statement
- In the ‘Research Proposal’ section of the application enter the name of the project you are applying to and upload your Individual research proposal, as mentioned above in BOLD
- In the funding field of your application, insert “I am applying for 2022 PhD Scholarship in Computer Science and Informatics”, and specify the project title and supervisors of this project in the text box provided.
- Qualification certificates and Transcripts
- References x 2
- Proof of English language (if applicable)
Interview - If the application meets the entrance requirements, you will be invited to an interview
If you have any questions or need more information, please contact [Email Address Removed]