PAN@FIRE: Overview of the PR-SOCO Track on Personality Recognition in SOurce COde

摘要

Author profiling consists of predicting an author's demographics (e.g. age, gender, personality) from her writing. After addressing at PAN@CLEF mainly age and gender identification, and also personality recognition in Twitter (http://pan.webis.de/), in this PAN@FIRE track on Personality Recognition from SOurce COde (PR-SOCO) we have addressed the problem of predicting an author's personality from her source code. In this paper, we analyse 48 runs sent by 1.1 participants. Given a set of source codes written in Java by students who answered also a personality test, participants had to predict big five traits. Results have been evaluated with two complementary measures (RMSE and Pearson product-moment correlation) that have allowed to identify whether systems with low error rates may work due to random chance. No matter the approach, openness is the trait that allowed to obtain the best results for both measures.