Abstract News articles are an engaging type of on-
line content that captures the attention of a significant
amount of Internet users. They are particularly enjoyed
by mobile users and massively spread through online so-
cial platforms. As a result, there is an increased inter-
est in discovering the articles that will become popular
among users. This objective falls under the broad scope
of content popularity prediction and has direct impli-
cations in the development of new services for online
advertisement and content distribution. In this paper,
we address the problem of predicting the popularity of
news articles based on user comments. We formulate
the prediction task as a ranking problem, where the
goal is not to infer the precise attention that a content
will receive but to accurately rank articles based on
their predicted popularity. Using data obtained from
two important news sites in France and Netherlands,
we analyze the ranking effectiveness of two prediction
models. Our results indicate that popularity prediction
methods are adequate solutions for this ranking task
Alexandru Tatar
LIP6/CNRS – UPMC Sorbonne Universit ́es
4 Place Jussieu, 75005, Paris, France
E-mail: tatar@npa.lip6.fr
Panayotis Antoniadis
Communication Systems Group – ETH Zurich
35 Gloriastrasse, 8092, Zurich, Switzerland
E-mail: antoniadis@tik.ee.ethz.ch
Marcelo Dias de Amorim
LIP6/CNRS – UPMC Sorbonne Universit ́es
4 Place Jussieu, 75005, Paris, France
E-mail: amorim@npa.lip6.fr
Serge Fdida
LIP6/CNRS – UPMC Sorbonne Universit ́es
4 Place Jussieu, 75005, Paris, France
E-mail: sf@npa.lip6.fr
line content that captures the attention of a significant
amount of Internet users. They are particularly enjoyed
by mobile users and massively spread through online so-
cial platforms. As a result, there is an increased inter-
est in discovering the articles that will become popular
among users. This objective falls under the broad scope
of content popularity prediction and has direct impli-
cations in the development of new services for online
advertisement and content distribution. In this paper,
we address the problem of predicting the popularity of
news articles based on user comments. We formulate
the prediction task as a ranking problem, where the
goal is not to infer the precise attention that a content
will receive but to accurately rank articles based on
their predicted popularity. Using data obtained from
two important news sites in France and Netherlands,
we analyze the ranking effectiveness of two prediction
models. Our results indicate that popularity prediction
methods are adequate solutions for this ranking task
Alexandru Tatar
LIP6/CNRS – UPMC Sorbonne Universit ́es
4 Place Jussieu, 75005, Paris, France
E-mail: tatar@npa.lip6.fr
Panayotis Antoniadis
Communication Systems Group – ETH Zurich
35 Gloriastrasse, 8092, Zurich, Switzerland
E-mail: antoniadis@tik.ee.ethz.ch
Marcelo Dias de Amorim
LIP6/CNRS – UPMC Sorbonne Universit ́es
4 Place Jussieu, 75005, Paris, France
E-mail: amorim@npa.lip6.fr
Serge Fdida
LIP6/CNRS – UPMC Sorbonne Universit ́es
4 Place Jussieu, 75005, Paris, France
E-mail: sf@npa.lip6.fr
and could be considered as a valuable alternative for
automatic online news ranking.
automatic online news ranking.