INAOE's Participation at PAN'13: Author Profiling Task Notebook for PAN at CLEF 2013

This paper describes the participation of the Laboratory of Language Technologies of INAOE at PAN 2013 evaluation lab. We adopted second order representations for facing the problem of Author Profiling (AP). This representation tackles two shortcomings of the typical Bag-of-Terms: i) the sparsity and high dimensionality of document representations, and ii) the assumption of total independence between terms in documents. In order to overcome these problems the proposed representation builds document vectors in a space of the different profiles, which represent the relationships of each document with the different profiles (say, age and gender). In order to evaluate our approach, we compare the proposed representation against a standard Bag-of-Terms representation using the PAN 2013 corpus for AP. We found that the second order attributes using a low computational cost, show evidence of being useful to determine genre and age profile.