Skip to main content

Featured Article: Estimating educational outcomes from students’ short texts on social media

An innovative model approach is proposed to predict the academic performance of students based on their posts on the popular Russian social networking site VK. A combination of unsupervised learning of word embeddings on a significantly large dataset of VK posts and a simple supervised model trained on individual posts is used. The model is able to differentiate posts from high- and low-performing students with an accuracy of almost 94%. The same model is further tested to predict the rankings of more than 900 high schools and the 100 largest universities in Russia based on their students' posts on VK. The ability to estimate and understand educational outcomes from simple social media posts could possibly help to determine unknown factors responsible for low or high academic performance. 

Submissions using our TeX template are strongly preferred. Please consider submitting your article in this format.

Articles

2018

Individual and Collective Human Mobility: Description, Modelling, Prediction
Edited by: Filippo Simini, Gourab Ghoshal, Luca Pappalardo, Michael Szell, Philipp Hövel


2016

Advances in data-driven computational social sciences
Edited by: Fosca Giannotti, Santo Fortunato, Michael Macy


2015

Making Big Data work: Smart, Sustainable, and Safe Cities
Edited by: Fabrizio Antonelli, Bruno Lepri, Alex 'Sandy' Pentland, Fabio Pianesi


2014

Scientific Networks and Success in Science
Edited by: Frank Schweitzer

Collective Behaviors and Networks
Edited by: Giovanni Luca Ciampaglia, Emilio Ferrara, Alessandro Flammini

Your browser needs to have JavaScript enabled to view this video

Your browser needs to have JavaScript enabled to view this video

Latest Tweets

Your browser needs to have JavaScript enabled to view this timeline

Latest article collections

Individual and Collective Human Mobility: Description, Modelling, Prediction
Edited by: Filippo Simini, Gourab Ghoshal, Luca Pappalardo, Michael Szell, Philipp Hövel

Advances in data-driven computational social sciences
Edited by: Fosca Giannotti, Santo Fortunato, Michael Macy

Making Big Data work: Smart, Sustainable, and Safe Cities
Edited by: Fabrizio Antonelli, Bruno Lepri, Alex 'Sandy' Pentland, Fabio Pianesi

See all article collections here

Aims and scope

The 21st century is currently witnessing the establishment of data-driven science as a complementary approach to the traditional hypothesis-driven method. This (r)evolution accompanying the paradigm shift from reductionism to complex systems sciences has already largely transformed the natural sciences and is about to bring the same changes to the techno-socio-economic sciences, viewed broadly.

EPJ Data Science offers a publication platform to address this evolution by bringing together all academic disciplines concerned with the same challenges:

    • how to extract meaningful data from systems with ever increasing complexity
    • how to analyse them in a way that allows new insights
    • how to generate data that is needed but not yet available
    • how to find new empirical laws, or more fundamental theories, concerning how any natural or artificial (complex) systems work

This is accomplished through experiments and simulations, by data mining or by enriching data in a novel way. The focus of this journal is on conceptually new scientific methods for analyzing and synthesizing massive data sets, and on fresh ideas to link these insights to theory building and corresponding computer simulations. As such, articles mainly applying classical statistics tools to data sets or with a focus on programming and related software issues are outside the scope of this journal.

EPJ Data Science covers a broad range of research areas and applications and particularly encourages contributions from techno-socio-economic systems, where it comprises those research lines that now regard the digital “tracks” of human beings as first-order objects for scientific investigation. Topics include, but are not limited to, human behavior, social interaction (including animal societies), economic and financial systems, management and business networks, socio-technical infrastructure, health and environmental systems, the science of science, as well as general risk and crisis scenario forecasting up to and including policy advice.

About EPJ

EPJ is a rapidly growing series of internationally reputed, peer-reviewed journals that are indexed in all major citation databases. The editorial boards of the EPJ are composed of leading specialists in their respective fields and have made it their mission to uphold the highest standards of scientific quality in the journals. EPJ started in the late 1990s as a merger and co-publication of Zeitschrift für Physik (Springer), Journal de Physique (EDP Sciences) and Il Nuovo Cimento (Società Italiana di Fisica) covering all aspects of the pure and applied physical sciences. Its spectrum has since expanded to encompass many interdisciplinary topics, including complexity and data sciences.

Co-publishers

EPJ Data Science is co-published by:

        

  


Annual journal metrics

Institutional membership

Visit the membership page to check if your institution is a member and learn how you could save on article-processing charges (APCs).

Funding your APC

​​​​​​​Open access funding and policy support by SpringerOpen​​

​​​​We offer a free open access support service to make it easier for you to discover and apply for article-processing charge (APC) funding. Learn more here