An extension of the Aspect PLSA Model to Active and Semi-supervised Learning for Text Classification


Anastasia Krithara(1), Massih-Reza Amini(3), Jean-Michel Renders(2), Cyril Goutte(3),
(1) NCSR Demokritos             (2)Xerox Research Center Europe                 (3) CNRC
    Patriarchou Grigoriou and Neapoleos         6, Chemin de Maupertuis         283, Bd Alexandre-Taché
                      Athens, Greece                                38240 Meylan                      Gatineau, QC J8X 3X7


In this paper, we address the problem of learning aspect models with partially labeled examples. We propose a method which benefits from both semi-supervised and active learning frameworks. In particular, we combine a semi-supervised extension of the PLSA algorithm with two active learning techniques. We perform experiments over four different datasets and show the effectiveness of the combination of the two frameworks.