Semi-supervised Document Classification with a Mislabeling Error Model

Anastasia Krithara(2), Massih-Reza Amini(1), Jean-Michel Renders(2), Cyril Goutte(3),
(1) Laboratoire d'Informatique Paris 6              (2)Xerox Research Center Europe                 (3) CNRC
                  104, Avenue du Président Kennedy                        6, Chemin de Maupertuis                  283, Bd Alexandre-Taché
                            75016 Paris                                                       38240 Meylan                            Gatineau, QC J8X 3X7

This paper investigates a new extension of the Probabilistic Latent Semantic Analysis (PLSA) model for text classification where the training set is partially labeled. The proposed approach iteratively labels the unlabeled documents and estimates the probabilities of its labeling errors. These probabilities are then taken into account in the estimation of the new model parameters before the next round. Our approach outperforms an earlier semi-supervised extension of PLSA introduced by which is based on the use of fake labels. However, it maintains its simplicity and ability to solve multiclass problems. In addition, it gives valuable information about the most uncertain and difficult classes to label. We perform experiements over the 20Newsgroups, WebKB and Reuters document collections and show the effectiveness of our approach over two other semi-supervised algorithms applied to these text classification problems.