Predicting age and gender in online social networks

By Claudia Peersman, Walter Daelemans, L. V. Vaerenbergh. SMUC '11, 2011

312 citations recorded by the source corpus.

A common characteristic of communication on online social networks is that it happens via short messages, often using non-standard language variations. These characteristics make this type of text a challenging text genre for natural language processing. Moreover, in these digital communities it is easy to provide a false name, age, gender and location in order to hide one's true identity, providing criminals such as pedophiles with new possibilities to groom their victims. It would therefore be useful if user profiles can be checked on the basis of text analysis, and false profiles flagged for monitoring. This paper presents an exploratory study in which we apply a text categorization approach for the prediction of age and gender on a corpus of chat texts, which we collected from the Belgian social networking site Netlog. We examine which types of features are most informative for a reliable prediction of age and gender on this difficult text type and perform experiments with different data set sizes in order to acquire more insight into the minimum data size requirements for this task.

Subfields: Artificial Intelligence, Human-Computer Interaction

View Paper: https://www.semanticscholar.org/paper/bdbc883649b90e04cfa3d18cea31023a22833de8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Human-Computer Interaction

Citations in the corpus

Key Female Characters in Film Have More to Talk About Besides Men: Automating the Bechdel Test (2015, background): “There has been much work in the computational sciences community on studying gender differences in the way language is used by men versus women (Peersman et al., 2011; Mohammad and Yang, 2011; Bamman et al., 2012; Schwartz et al., 2013; Bam-man et al., 2014; Prabhakaran et al., 2014).”
Cybercrime detection in online communications: The experimental case of cyberbullying detection in the Twitter network (2016, background): “Note that as mentioned above, most OSN users do not clearly state their age and gender in their online profiles (Peersman et al., 2011); thus, many studies (Hosseini & Tammimy, 2016; Peersman et al.”
Zooming in on Gender Differences in Social Media (2016, background): “In (Peersman et al., 2004), age and gender prediction is performed on short messages from social networking sites.”
CLiPS Stylometry Investigation (CSI) corpus: A Dutch corpus for the detection of age, gender, personality, sentiment and deception in text (2014, background): “At the moment, there exist a number of Dutch corpora for the detection of age, gender (Peersman et al., 2011; Nguyen et al., 2013), authorship and personality (Luyckx and Daelemans, 2008).”
Computational Sociolinguistics: A Survey (2016, methodology): “Unigrams alone already perform well [185, 186, 198].”
Semi-supervised Gender Classification with Joint Textual and Social Modeling (2016, background): “As for supervised learning, gender classification has been extensively studied in several textual styles, such as Blog (Nowson and Oberlander, 2006; Peersman et al. , 2011; Gianfortoni et al. , 2011), E-mail (Mohanmad et al. , 2011), YouTube (Filippova, 2012) and Micro-blog (Rao et al. , 2010; Liu…”
Why Gender and Age Prediction from Tweets is Hard: Lessons from a Crowdsourcing Experiment (2014, methodology): “, 2005), telephone conversations (Garera and Yarowsky, 2009), YouTube (Filippova, 2012) and chats in social networks (Peersman et al., 2011).”
Overview of the Author Profiling Task at PAN 2013 (2013, methodology): “[25] who retrieved a dataset from Netlog2,where authors report their gender and exact age, and Koppel et al.”
User Classification with Multiple Textual Perspectives (2016, methodology): “Peersman et al. (2011) apply a text categorization approach to age classification with textual features only, i.e., word unigrams and bigrams.”
Overview of the 3rd Author Profiling Task at PAN 2015 (2015, background): “Irrespective of the good performance of n-gram features as reported in [27] and [52], it is worth to mention the effect of more elaborated features as well.”
Gender Profiling for Slovene Twitter communication: the Influence of Gender Marking, Content and Style (2017, background): “In the last few years, more languages have received attention in the context of author proﬁling (Peersman et al., 2011; Nguyen et al., 2013; Rangel et al., 2015; Rangel et al., 2016), with the publication of the TwiSty corpus containing gender information on Twitter authors for six languages…”
External validity of sentiment mining reports: Can current methods identify demographic biases, event biases, and manipulation of reviews? (2014, unknown): “The list of papers found is here: [12,14–17,26,28,31–35,43–60].”