AAA Data Science Postdoctoral researcher on Quality, Trust and Reliability of Web Data

Web & Media Group (VU) & Department of Media Studies (UvA)

VU University Amsterdam is one of the leading institutions for higher education in Europe and aims to be inspiring, innovative, and committed to societal welfare. It compromises twelve faculties and has teaching facilities for 25.000 students.

The Faculty of Sciences performs research in the areas Life & Health, Networked World, Fundamentals of Science, and Energy & Sustainability. More than 1700 students participate in mono- and interdisciplinary bachelor and master courses. The staff is hosted in the departments of Physics, Chemistry & Pharmaceutical Sciences, Mathematics, and Computer Sciences. 

Amsterdam Data Science, an initiative of the University of Amsterdam (UvA), VU University Amsterdam (VU), Amsterdam University of Applied Sciences (HvA), and Centrum Wiskunde & Informatica (CWI) is looking for 14 researchers at the postdoctoral/PhD level. These positions are funded by the Amsterdam Academic Alliance (AAA), a joint initiative of the UvA and VU aimed at intensifying collaboration with each other and knowledge institutions in the region, to cement Amsterdam's position as a hub of academic excellence. These positions are partially co-funded by CWI, Faculty of Social and Behavioural Sciences of the UvA, HvA, ORTEC, Spinoza fund of Prof. Vossen, and VUmc.

Amsterdam Data Science provides a network in which some 300 Amsterdam based scientists exchange Data Science knowledge and forge partnerships within and outside the academic community. This comprises the foundations of data science and data science for the life sciences, social analytics, business analytics, and digital humanities.

This advertisement concerns one of the 14 positions. The Network Institute of the VU University of Amsterdam and the Amsterdam School for Heritage and Memory Studies at the University of Amsterdam are looking for a motivated PhD candidate or postdoctoral researcher for the project 'Representation of Data Quality', in the context of a larger research program called 'QuPiD2: Quality and Perspectives in Deep Data'. The candidate will be part of a collaborative team aiming all together to achieve a formal modeling of quality and perspectives. The candidate will be also part of the Network Institute research network at the VU University Amsterdam and the Amsterdam School for Heritage and Memory Studies and will work within multidisciplinary teams of humanities researchers and computer scientists.

Since the Web is tied in with freedom of speech, Web data are diverse and biased, which limits their uptake as sources for Humanities research. Evaluating the reliability of these 'noisy' data at the scale of the Web requires contextualizing the provenance of the data and analyzing their reflected perspectives, as well as a comparison with information extracted from trusted data repositories. This project aims to define measures for trust and reliability in historical textual data by combining NLP-processing, crowdsourcing human annotations and social media analysis.

Project description

As part of the QuPiD2 research team, the candidate will

1. investigate the identification of factors that influence the quality of textual sources, such as provenance and the perspectives they reflect; and
2. provide grounding for the reliability of multimedia content on the Web. The results of this research will be tested on a shared dataset across the QuPiD2 program, including newspapers, social media, biographies, encyclopedias, literary texts. The outcome should serve as the basis for services that promote informed decision making in the fields of consumption, health, politics and education.

Tasks: 

• Study existing approaches to data quality, reliability and trust on the web;
• study the role of provenance and perspectives in data quality, reliability and trust on the web;
• study the specific data sources and use cases related to the QuPiD2 program. Both English and Dutch sources could be considered;
• develop an approach to determine features that influence data quality, and scalable measures for trust and reliability in historical textual data by combining NLP-processing, crowdsourcing human annotations and social media analysis;
• using the machine-crowd empowered processing of textual and social media sources for populating QuPiD2 model;
• create datasets for training and evaluation through domain expert and crowd annotations;
• perform continuous performance evaluation for the application use cases, using the evaluation datasets;
• collaboration with the QuPiD2 program research team;
• publish the results of the work as scientific articles in high ranked journals and conferences, as well as present the work at relevant scientific venues.

Requirements

The candidate should have a strong background (MSc, PhD) in computer science and/or human computing, machine learning, web technologies, linked data, and a strong affinity with methodological issues in the humanities, in particular historical source criticism. The candidate should have sufficient programming skills and experience with data modelling, engineering and mining, as well as affinity to work in multi-disciplinary team.

Further information

For additional information please contact:

• Dr. Lora Aroyo
  T: +31 (0)20 620329972
• Prof. Julia Noordegraaf
  T: +31 (0)20 5257279

• Working at VU

Appointment

The initial appointment will be for a period of three years. We have excellent employment conditions such as:

• remuneration of 8,3% end-of-year bonus and 8% holiday allowance;
• solid pension scheme (ABP);
• a minimum of 29 holidays In case of full-time employment;
• possibilities to save up holidays for sabbatical leave.

The salary will be in accordance with university regulations for academic personnel, and depending on experience, range from a minimum of €2,476 gross per month up to a maximum of €3,908 gross per month (salary scale 10) based on a full-time employment. The Collective Labour Agreement for Dutch Universities is applicable.

Job application

Applications may only be submitted via informatica.secretariaat.few@vu.nl. To process your application immediately, please quote vacancy number 15111 and the title of the position you are applying for in the subject-line.

Applications must include a detailed curriculum vitae, BSc and MSc academic transcripts, a motivation letter explaining why you are the right candidate, list of projects you have worked on with brief descriptions of your contributions and the names and contact addresses of two academic references from which information about the candidate can be obtained. All these should be grouped in one PDF attachment.

Applications will be accepted until 10 May 2015.

No agencies please