[HPSG-L] PhD position in Data-Driven Computational Semantics and Geographic Information Systems

[Yoad Winter]

PhD position in Data-Driven Computational Semantics and Geographic 
Information Systems (1.0 FTE)

Deadline: 17 June 2018

The Utrecht Institute of Linguistics (UiL OTS) invites applications for 
a PhD position in data-driven computational semantics. The project will 
use large corpora and machine learning methods to acquire meanings of 
spatially-based concepts, which will be used as part of a natural 
language interface to Geographic Information Systems (GIS). The project 
will enrich an initial model of spatial meanings (Kracht 2002, Mani & 
Pustejovsky 2012, Zwarts & Winter 2000) with a selected set of semantic, 
syntactic, and pragmatic parameters. To train this model, the project 
will acquire a large dataset of language use with spatial language, 
including ordinary texts and expert interactions with GIS. These data 
will be employed for training and evaluating the initial model. This 
novel combination of semantic representation, data-driven learning and 
evaluation against a real-life problem is expected to substantially 
advance the ability of users to interact with geographic databases.
The project will be supervised by Dr. Tejaswini Deoskar and Prof. Yoad 
Winter, in collaboration with Dr. Joost Zwarts and NLP researchers from 
the Language and Computation group at the ILLC, University of Amsterdam 
(Prof. Khalil Sima’an, Dr. Jelle Zuidema). The empirical study of 
elicited user data will be carried out in collaboration with Prof. James 
Hampton (Psychology, City, University of London). Work at the UiL OTS 
will be part of the ROCKY project, financed by an ERC Advanced grant to 
Prof. Yoad Winter.

References
Kracht, M., 2002. On the semantics of locatives. Linguistics and 
Philosophy, 25(2), pp.157-232.
Mani, I. and Pustejovsky, J., 2012. Interpreting motion: Grounded 
representations for spatial language. Oxford U.P.
Zwarts, J. and Winter, Y., 2000. Vector space semantics: A 
model-theoretic analysis of locative prepositions. Journal of logic, 
language and information, 9(2), pp.169-211.


Qualifications
The successful applicant for this position should have a strong 
background in artificial intelligence, machine-learning and computer 
science as well as an interest in formal semantics. Applicants should 
ideally hold, or be about to complete, an excellent Master's degree in 
Computational Linguistics or a related domain (Artificial Intelligence, 
Machine Learning and/or Computer Science). Good programming skills, 
proficiency in spoken and written English, as well as a keen interest in 
problem solving and research are essential.


Application details and additional information
Look up the job ad at:
https://www.uu.nl/en/organisation/working-at-utrecht-university/jobs
For more information, please contact the Project Director (Yoad Winter) 
or the Research Coordinator at UiL OTS (Maaike Schoorlemmer).

Apply online through:
https://www.uu.nl/en/organisation/working-at-utrecht-university/jobs

The application deadline is
17/06/2018

Interviews will be held during the week starting June 25, 2018.