Postdoctoral Researcher/Scientific Programmer in the field of Social Sciences and Humanities (SSH)

Vacancy number 20-319

Function type Academic staff

Hours (in fte) 1.0

External/ internal External

Location Leiden

Placed on 21 July 2020

Closing date 31 July 2020

The Faculty of Humanities, Leiden University Centre for Linguistics (LUCL) is looking for a

Postdoctoral Researcher/Scientific Programmer in the field of Social Sciences and Humanities (SSH) 

Project description
As a scientific programmer, you will join a broad consortium of researchers (Leiden University, Utrecht University, Radboud University Nijmegen, Erasmus University Rotterdam, University of Groningen) in a digital infrastructure project (PDI-SSH funding). The aim of the project is to create the first set of deep neural language models pre-trained on historical textual material (Dutch and English) from different time periods. The impressive performance of neural language models trained on Present-day languages has amply been demonstrated in a wide range of NLP tasks, many of which are of substantial value to SSH research (part-of-speech tagging, lemmatization, coreference resolution, sense-disambiguation, dependency parsing, semantic role labelling, paraphrase detection, sentiment analysis, textual entailment analysis, wikification). The research infrastructure that will result this project will support text-oriented SSH research by extending the number of possibilities for searching in formally digitized text corpora (e.g. onomasiological searches), and facilitating (semi)-automatic data retrieval, annotation and analysis. At the same time, the research team will explore the potential of employing neural language models to address theoretical questions about the nature of social and cultural change.

Key responsibilities
The tasks of the Postdoctoral Researcher/Scientific Programmer are twofold. In close collaboration with the project leaders, your will design and develop the research infrastructure, which will be made available to SSH researchers and other interested users through an open-source software package and notebooks for building Python programs to use the pre-trained models of historical Dutch and English. This package provides a suite of libraries to access and fine-tune the models, create embeddings for the researcher’s specific data set, and evaluate and visualize results. Specific tasks include:

• Determining quantity and quality of resources and pre-processing the training data: delineating genre balance & time periods of training data, normalization, foreign-language detection;
• Exploring different model architectures and tokenizers: Wordpiece vs. Byte-Pair Encoding tokenizer;
• Experimenting with different training objectives: Next Sentence Prediction, Masked Language Modelling, and Sentence Order Prediction;
• Technical evaluation through different downstream tasks: genre classification, POS-tagging;
• Creating instructional notebooks and evaluating of models by means of domain expert feedback;
• Disseminating the Open Source materials via workshops, presentations and publications.

At the same time, you will closely collaborate with domain experts in Historical (socio-)linguistics, Literary and Cultural Studies, and Social Sciences on innovative case studies.

Selection criteria

• PhD degree in Computational Linguistics;
• Ample experience with text mining in non-standard and/or historical language data;
• A broad interest in computational approaches to text-based Humanities (Literary Studies, Linguistics) and Social Sciences;
• Experience with designing (Open Source) Software packages;
• Excellent programming skills in the following languages: Python, Java;
• Familiarity with: Bidirectional LSTM and/or Transformer models, jupyter notebooks, R;
• An excellent command of English, good command of Dutch.

Our organisation
The Faculty of Humanities is rich in expertise in fields such as philosophy, religious studies, history, art history, literature, linguistics and area studies covering nearly every region of the world. With its staff of 995, the faculty provides 27 master’s and 25 bachelor’s programmes for over 7,000 students based at locations in Leiden and in The Hague.

The Faculty has seven Institutes, among which the Leiden University Centre for Linguistics (LUCL). LUCL has a longstanding tradition in the study of the world’s languages and features unique linguistic expertise. Current theoretical insights are combined with modern experimental methods in its research profile area ‘Language Diversity in the World’.

Terms and conditions
We offer a full-time temporary appointment of two years. Salary in the first year € 4,012.- gross per month, in the second year € 4,139.- gross per month (pay scale 11, step 2 and 3, in accordance with the Collective Labour Agreement for Dutch Universities). Depending on qualifications, the researcher may start at the appropriate step in scale 10 until the candidate fully meets the requirements for scale 11 as specified by the Faculty of Humanities, particularly with regard to the number of years of relevant work experience. The intended starting date is 1 April 2021.

Leiden University offers an attractive benefits package with additional holiday (8%) and end-of-year bonuses(8.3 %), training and career development and sabbatical leave. Our individual choices model gives you some freedom to assemble your own set of terms and conditions. For international spouses we have set up a dual career programme. Candidates from outside the Netherlands may be eligible for a substantial tax break. More information can be found at the website.

Diversity
Leiden University is strongly committed to diversity within its community and especially welcomes applications from members of underrepresented groups.

Information
Enquiries can be made to Lauren Fonteyn, email l.fonteyn@hum.leidenuniv.nl.

Applications
Please submit online your application no later than 31 July 2020 via the blue button in our application system. Your application should include:

• Curriculum Vitae and list of publications;
• A cover letter explaining your motivation, background and qualifications for the position (max. 1 page).