Personality facets recognition from text

Fundamental Big Five personality traits and their facets are known to correlate with a wide range of linguistic features and, accordingly, the recognition of personality traits from text is a well-established NLP task. Obtaining facets information may however require extensive personality inventories and, as a result, existing computational models are usually limited to the recognition of the five main personality categories. Based on these observations, this paper investigates the recognition of a number of personality facets from a Brazilian Facebook corpus obtained (at low cost) from a shorter personality inventory. In doing so, we compare a number of personality facet recognition models, and present preliminary reference results for further studies in the field.