myPersonality Project

Personality Recognition consists in the automatic classification of authors’ personality traits, that can be compared against gold standard annotation obtained by means of the big5 personality test.
In recent years the interest of the scientific community towards Personality Recognition has grown incredibly, since the application of Personality Recognition proved to be a valuable resource in Social Network Analysis and Social Computing.
Despite there are a growing number of works in Personality Recognition, it is still very difficult to say what is the state-of-the-art, because almost all the scholars working in the field used different datasets and evaluation measures.
In the Workshop on Computational Personality Recognition (Shared Task) we are going to run a shared task that will allow participants to compare the results of their systems on a common benchmark. We do not want to focus just on performance, but rather on discovering which feature sets, resources and learning techniques are useful in the extraction of personality from text. The workshop will be held in Boston on July 11 2013, in conjunction with ICWSM. Organizers are going to let the data available as a benchmark for future experiments, after the shared task.

The Workshop on Computational Personality Recognition (Shared Task) invites contributions from researchers or teams working in:
-Data Mining,
-Computational Psychology,
-Natural Language Processing,
-Social Network Analysis,
-Sentiment Analysis,
-Deception Detection,
-Information Extraction,
-Human-Computer Interaction,
-Authorship Attribution
and any other related area.

Organizers provide two gold standard labelled datasets (released on Feb 1st, 2013):

1. essays.zip: a large dataset of 2400 stream-of-consciousness texts labelled with personality, produced by Pennebaker & King 1999 and used by Mairesse et al. 2007.
   
2. mypersonality_final.zip: About 10000 Facebook status updates of 250 users, plus Facebook network properties (including network size, betweenness centrality, density and transitivity) labelled with personality, produced by Celli, Pianesi, Stillwell & Kosinski, 2013. **NOTE: mypersonality dataset has been updated on feb, 15, 2013. if you downloaded it before that date please download it again.

Contributors are required to:

• use at least one of the datasets provided by the organizers for their experiments (more info in the task guidelines.zip);
  
• provide the files used for the experiments;
  
• submit a short paper (max. 4 pages, including references) reporting all the information about features, resources and techniques used in the experiments, and discussing results.

Organizers warmly invite contributors working in Personality Recognition to test their systems on the datasets released. Other researchers or teams that wish to use the datasets for the tasks listed above are more than welcome.
The Programme committee will evaluate and select the papers for publication on the basis of clarity, correctness, meaningful comparison, significance, soundness and replicability. Negative results will be considered as important as positive results.
Selected papers will be published in the workshop proceedings and will be presented as posters at the workshop.
All papers must be anonymous and follow AAAI formatting guidelines. Papers must be sent to fabio.celli@live.it no later than March 20th, 2013 (deadline extended).
The Workshop will take place in a half day on July 11th, 2013 and will include a presentation talk, invited talks, a poster session and a panel discussion, open to contributors, organizers and attendees.

January 21th, 2013: first call for participation
January 28th, 2013: second call for participation
February 1st, 2013: data release
March 20th, 2013: paper due date (extended deadline)
March 25th, 2013: notification of acceptance
March 29th, 2013: camera-ready deadline
July 11th, 2013: Workshop.

Preface:

• Celli F., Pianesi F., Stillwell D. and Kosinski M. Workshop on Computational Personality Recognition (Shared Task) pdf
  

Papers:

• Verhoeven B., Daelemans W. and De Smedt T. Ensemble Methods for Personality Recognition pdf
  
• Farnadi G., Zoghbi S., Moens M.F. and De Cock M. Recognising Personality Traits using Facebook Status Updates pdf
  
• Tomlinson T.M., Hinote D. and Bracewell D.B. Predicting Conscientiousness through Semantic Analysis of Facebook Posts pdf
  
• Markovikj D, Gievska S., Kosinski M. and Stillwell D. Mining Facebook Data for Predictive Personality Modeling pdf
  
• Alam F., Stepanov Evgeny A. and Riccardi G. Personality Traits Recognition on Social Network - Facebook pdf
  
• Mohammad S. and Kiritchenko S. Using Nuances of Emotion to Identify Personality pdf
  
• Appling D.S. Briscoe E.J., Hayes H. and Mappus R.L. Towards Automated Personality Identification using Speech Acts pdf
  
• Iacobelli F. and Culotta A. Too Neurotic, not too Friendly: Structured Personality Classification on Textual Data pdf
  

9:00am - 9:30am > François Mairesse (keynote speaker) From Personality Recognition to User Modeling: Moving Towards Task-Based Evaluation
9:30am - 10:00am > Daniel Gatica-Perez (keynote speaker) Big-Five in 100 million pixels: inferring personality in social video
10:00am - 10:30am > Fabio Celli (organizer) WCPR13: results and next challenges
10:30am - 10:45am > Break
10:45am - 11:45am > Poster Session
11:45am - 12:30pm > Panel Discussion

François Mairesse. Nuance Communications.
Cindy Chung. University of Texas, Austin.
Gianluca Lebani. University of Trento.
Bruno Lepri. Massachussetts Institute of Technology.
Ailbhe Finnerty. Fondazione Bruno Kessler.
Tommaso Fornaciari. University of Trento.
Marco Milano. Fondazione Bruno Kessler.
Scott Nowson. Appen Butler Hill.
Luca Rossi. University of Urbino.
Kyriaki Kalimeri. Fondazione Bruno Kessler.
Francesca Bonin. Trinity College Dublin.
Matthias Mehl. University of Arizona.
Elia Bruni. University of Trento.
Daniel Gatica-Perez. Polytechnique de Lausanne.
Alastair Gill. King's College, London.

Fabio Celli, Center for Mind and Brain Sciences, University of Trento.
Fabio Pianesi Fondazione Bruno Kessler.
David Stillwell, Psychometrics Centre, University of Cambridge.
Michal Kosinski, Psychometrics Centre, University of Cambridge.

Confidentiality
You cannot send the database to any other party, nor disclose to anyone else the information contained within it as well as its structure.

Anonymisation
We have anonymised the dataset. You cannot attempt to reverse this by linking individual records with specific users. This means that you also should not link individual data with any other information about an individual that you may have. This implies that you cannot attempt to contact any individuals either.

Non-Commerical License
We grant you a non-commercial license to use the data. You can only use it for academic research that does not earn revenue, and your research also cannot be in collaboration with any commercial entities.

Recoverability
The license to use and store our data is recoverable, which means that myPersonality may ask you to cease use of it and to delete it from any storage you have at our sole discretion.

This page is managed by Fabio Celli