[Corpora-List] CFP: 2nd Workshop on Computational Personality Recognition (WCPR14)

[fabio celli]

**Sorry for cross-posting**

Call for Papers/Participation to shared task and Competition
The Workshop on Computational Personality Recognition 2014 is a challenge-based shared task at its second edition and this year will be held in conjunction to ACMMM in Orlando (FL) on novemeber 7. The main goals of the workshop is to define the state-of-the-art and release data/tools for a standard evaluation of computational personality recognition. Organizers warmly invite contributors working inPersonality RecognitionData Mining, Computational Psychology, Natural Language Processing, Social Network Analysis, Sentiment Analysis,Opinion Mining,Mood Detection, Deception Detection, Information Extraction, Human-Computer Interaction, Authorship Attribution,and other related areas,to submit papers and participate to the competition.Task organizers will provide an introductory paper to summarize the results and describe the task. The results of the workshop will be available to the public. 
Organizers released two gold standard labelled datasets and one evaluation tool:- Youtube personality dataset, described in Biel & Gatica-Perez 2013.- Mobile personality dataset, described in Aharoni et al. 2011; Staiano et al. 2012.- Evaluation tool.The  shared task is structured in two tracks: one open track and two competitions. 
Track 1a: competition on multimodal personality recognition. Contributors are required to use the Youtube personality dataset or the Mobile personality dataset to: 1) develop their own system for personality type classification, that generates the predictions in the format required by the evaluation tool; 2) submit a short paper (max. 4 pages, including references) indicating "multimodal personality recognition" in the keywords, and reporting all the information about features, resources and techniques used in the experiments. 3) report predictions file while submitting the paper. 
Track 1b: competition on personality recognition from text. Contributors are required to use only the text transcriptions from the Youtube personality dataset to: 1) develop their own system for personality type classification, that generates the predictions in the format required by the evaluation tool; 2) submit a short paper (max. 4 pages, including references) indicating "personality recognition from text" in the keywords, and reporting all the information about features, resources and techniques used in the experiments. 3) report predictions file while submitting the paper. 
Track 2: open shared task. Contributors are required to: 1) use at least one of the datasets provided by the organizers; 2) submit a short paper (max. 4 pages, including references) indicating "personality shared task" in the keywords, and discussing their results. Contributors are allowed to use any other corpus or resource, including the ones released in the previous edition of the workshop, to provide useful comparisons between personality recognition and other research fields.
more info on the workshop website https://sites.google.com/site/wcprst/home/wcpr14
Regards.
Fabio CelliPost DocDepartment of Information Engineering and Computer ScienceUniversity of Trentohttp://clic.cimec.unitn.it/fabio/ 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20140407/c830d532/attachment.htm>
-------------- next part --------------
_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora