Teamwork has become an important part of the educational process, aiming for preparing students for their future performance, but also for making the learning process easier. Predicting students’ performance in advance is one of the keys to prevent failure, but teamwork results are difficult to estimate and impair the global prediction results [1]. Working in group is an inherent social activity, and measuring students’ performance in it may be enhanced if understood in that context. In this paper, we propose to quantify the performance of students in teamwork by making use of the most effective techniques for social networks analysis. Teamwork is then represented as a network, where students interact with each other, achieving some results (their grades). We explore a network architecture and provide a strategy for quantifying the global contribution of each student through adaptations of the PageRank algorithm [2].
[1]
Charu C. Aggarwal,et al.
An Introduction to Social Network Data Analytics
,
2011,
Social Network Data Analytics.
[2]
Jon Kleinberg,et al.
The link prediction problem for social networks
,
2003,
CIKM '03.
[3]
Sergey Brin,et al.
The Anatomy of a Large-Scale Hypertextual Web Search Engine
,
1998,
Comput. Networks.
[4]
Jonathan L. Gross,et al.
Handbook of graph theory
,
2007,
Discrete mathematics and its applications.
[5]
Jon M. Kleinberg,et al.
The link-prediction problem for social networks
,
2007,
J. Assoc. Inf. Sci. Technol..
[6]
Cláudia Antunes,et al.
Anticipating student’s failure as soon as possible
,
2009
.