The research group Quantitative Lexicology and Variational Linguistics (http://wwwling.arts.kuleuven.be/qlvl/) of the Department of Linguistics at KU Leuven, Belgium invites submissions for three research positions on their newly initiated projects Nephological Semantics and MARS.
"Nephological Semantics. Using token clouds for meaning detection in variationist linguistics" addresses the question of how token-based vector semantics, as an advanced type of distributional meaning analysis, can be put to use in the linguistic study of semasiological and onomasiological variation, including applications in aggregate lectometrical analysis and grammatical alternation research. The project is supervised by Dirk Geeraerts, Dirk Speelman, Stefania Marzo and Benedikt Szmrecsanyi. For a more detailed description, see * Nephological Semantics * (http://wwwling.arts.kuleuven.be/qlvl/NephoSem.html).
"MARS. MAchine Reading of patient recordS" is a joint project of computer scientists, linguists, medical specialists and legal scholars, that addresses the fundamental problems for automatic information extraction from grammatically and lexically irregular language as it is used by medical professionals in hospital patient records. QLVL's contribution will focus on modelling the lexical and terminological variation through corpus-based and statistical analysis of recurrent, schematic and linguistically motivated variation patterns. The resulting insights into the dynamics of lexical variation will inform the future development of new computational approaches to Biomedical Named Entity Recognition in machine reading applications. Overall project supervisor is Marie-Francine Moens of the LIIR research group at the Computer Science department. QLVL supervisors are Ann Bertels and Kris Heylen. For a more detailed description, see * Machine Reading Patient Records * (http://wwwling.arts.kuleuven.be/qlvl/MARS.html).
* Postdoctoral position in distributional semantics (3 yrs) *
Your task will be to develop and finetune Semantic Vector Space-based tools for polysemy detection and semantic description. This task forms a central part of the Nephological Semantics project. You have completed or are about to complete a PhD in linguistics, computational linguistics, or a similar discipline, and you have demonstrable skills in working with corpus-based distributional methods and/or statistical approaches to Word Sense Disambiguation. You have a keen interest in natural language semantics and more specifically in the social, contextual and cognitive factors that co-determine the meaning of words in actual use. You are ready to work in a highly collaborative environment. You are fluent in spoken and written English. Knowledge of Dutch is an asset but not a requirement.
* PhD position in statistical modelling of terminological variation (2yrs + 2yrs renewable) *
Working on the interdisciplinary MARS project, you will do innovative research at the crossroads between the linguistic analysis of lexical variation and statistical NLP approaches to term extraction with an application in the medical domain. You will develop your basic computer programming skills for handling large and diverse collections of linguistic data. You will explore different state-of-the-art statistical analysis techniques to identify typical variation patterns in the medical terminology of patient records and generalize these to new insights in the dynamics of lexical variation. You have completed or are about to complete a Master in linguistics, computational linguistics, or a similar discipline. If your main background is in linguistics, you have additionally specialized, gained experience or displayed a strong interest in computational and statistical analysis of language data. If your main background is in computer science, you have additionally specialized, gained experience or displayed a strong interest in dealing with natural language phenomena. You are ready to work in an interdisciplinary and highly collaborative environment. You are fluent in spoken and written English. Knowledge of Dutch is an asset but not required at the time of application. However, since a substantial part of the data is in Dutch, you must be willing to acquire a practical reading knowledge during the project.
* Software engineer (2 yrs) *
Your task will be to work on a scalable high-performance software solution for data analytics applied to large bodies of free-text natural language data (particularly through semi-supervised feature learning, using context-sensitive token-based distributional semantics). More specifically, you will refine, optimize, and expand on an existing code base, addressing issues of scalability, parallelisation, data flow optimisation, and visualization of distributional patterns. The software that is to be developed will be used in both the Nephological Semantics project and the MARS project. You have a master's degree in informatics/computer science. You have an outspoken interest in big data, data mining and data analytics, and you have solid knowledge of database technology, data structures and algorithms for large data, and programming in C/C++, Java, and Python. Experience with visualisation tools such as D3.js, with scientific computing tools such as Octave or MATLAB, and with high-performance computing is considered an asset. Familiarity with the Linux operating system is required. Also, you are ready to work in an interdisciplinary and highly collaborative environment, and you are fluent in spoken and written English.
All three positions start in October 2015 at the earliest. Remuneration is in accordance with the official Belgian scales for academic personnel and bursaries, which offer internationally highly competitive pay and benefits to R&D professionals in the early stages of their careers (see the KU Leuven jobsite for more information: http://www.kuleuven.be/personeel/jobsite/en/working).
If you are interested in any of these positions, please send a statement of interest to qlvl at kuleuven.be no later than July 15, 2015. The letter should explain your motivation, and should contain a CV, a copy of your MA or PhD thesis (or a draft thereof), a copy of relevant publications (if any), and one or more letters of reference and/or the contact details of one or more referents (name, affiliation, email, telephone number), all bundled in one pdf.