June 20 to July 1, 2022 | ODISSEI - Erasmus University Rotterdam
The SICSS-ODISSEI will show participants all the possibilities that ODISSEI has to offer in terms of facilities, tools and research support. SICSS-ODISSEI will not only focus on teaching participants relevant methodologies, theories and techniques for computational social science, but also allow participants to apply this knowledge on ODISSEI facilities, including executing specific projects using microdata via the CBS Secure Remote Access Environment and an introduction to the ODISSEI Secure Supercomputer.
Participants will be able to develop research ideas with other computational social scientists and turn these into exciting new projects that can be executed within ODISSEI. Participants will be supported by experts in the field and projects could be developed into further long-term research, a grant application or scientific publication. At the end of the two weeks, participants will be able to present their research as a poster at the European Consortium of Sociological Research Conference which will be held in Amsterdam on 7-9 July.
Participants will get an introduction to working with CBS (Statistics Netherlands) data, where they’ll be able to see the full potential of administrative data, high performance computing and intensively linked data. Participants will also be guided on how to develop their ideas into large projects and learn how ODISSEI infrastructure can be adapted to their own needs. During the SICSS-ODISSEI programme, participants will get lectures from colleagues at ODISSEI member organisations on topics such as Network Analysis, Machine Learning, Geospatial Analysis and Benchmarking and apply these in hands-on workshops. During the SICSS-ODISSEI programme, participants will get lectures from colleagues at ODISSEI member organisations, such as SoDa, CBS, Centerdata, Netherlands Twin Register.
Participants will need to develop their knowledge and skills in a relatively short period, so in preparation for the summer school participants are able to follow a workshop in data carpentry, organised by the Netherlands eScience Center, to enhance their skills prior to SICSS-ODISSEI.
SICSS-ODISSEI will run from Monday 20 June to Friday 1 July 2022. Both weeks, from Monday to Friday, participants will learn about computational social science in-person in the Netherlands. The regular days of teaching will be at the campus of the Erasmus University of Rotterdam, Burgemeester Oudlaan 50. There may be some site visits to ODISSEI partners during the summer school, which will be confirmed at a later date.
During the first week, the following topics will be covered: Introduction to ODISSEI, Machine learning, Network Analysis, Benchmarking. Please see the schedule for more details.
In the second week of the summer school, participants will be able to make full use of the ODISSEI infrastructure through a full scale benchmarking challenge, during which they will be put into teams and challenged to work on and submit solutions to a real-life prediction problem that social scientists and policy makers are grappling with. An example topic would be to predict educational or labour market outcomes. For an example of such a social science benchmark that was organised a few years ago see the Fragile Families Challenge in the onboarding reading.
The exact challenge will only be revealed at the summer school. This challenge is however fairly unique as each team will be provided with access to Statistics Netherlands Microdata, containing detailed data on the whole dutch population. This is a treasure trove of data which can help us predict important outcomes. The teams will compete to develop the most effective predictive model for a range of specified target variables. This benchmarking challenge will help participants get a more hands-on experience of computational social science research with some of the most advanced data infrastructure available. This week will be led and supervised by dr. Paulina Pankowska and the benchmarking project team. ODISSEI’s Social Data Science Team will help participants overcome practical obstacles they may encounter while working on their project. All solutions submitted by the teams will be compared against the same ground truth data using the same evaluation metrics. Towards the end of the workshop the teams will get the chance to present their approach. The solutions will be assessed by a committee, which will take into account, among other things, the predictive performance of the proposed models, the narrative explanation of the approach, and its innovativeness. The winners of the challenge will be announced during the ECSR conference that will take place in Amsterdam in July, 2022.
Important information regarding CBS microdata access:
The ODISSEI infrastructure includes access to highly sensitive and secure data held at Statistics Netherlands. Statistics Netherlands operates strict security protocols that may limit the access of some participants to data used in the benchmarking challenge held in the second week of the summer school. Participants with an employment contract at a Dutch Research Organisation will be able to access the data at Statistics Netherlands subject to completing the necessary access requirements. If you are not employed at a Dutch Research organisation then, in partnership with Statistics Netherlands, we will assess if it is possible to provide you with access to the administrative data used in the benchmarking challenge. This will be done on a case by case basis. If this is not possible, we will work with you to ensure that you get the maximum out of the second week of the summer school possible and are included within the benchmarking challenge. If you would like to discuss this in more detail, please email firstname.lastname@example.org and we will happily discuss ways in which you can get the most of your time on SICSS-ODISSEI.
More information about the schedule and collaborations will be announced soon.
You can host a partner location of the Summer Institutes of Computational Social Science (SICSS) at your university, company, NGO, or government agency.