Détection automatique d'opinions dans des corpus de messages oraux
Nathalie Camelin, Géraldine Damnati, Frédéric Béchet, Renato De Mori.

Telephone surveys are often used by Customer Services to evaluate their clients' satisfaction and to improve their services. Large amounts of data are collected to observe the evolution of customers' opinions. Within this context, the automatization of the process of these databases becomes a crucial issue. This paper addresses the automatic analysis of audio messages where customers are asked to give their opinion over several dimensions about a Customer Service. Interpretation methods that integrate automatically and manually acquired knowledge are proposed. A set of classifiers with several input features is introduced for each type of knowledge to add robustness in reducing the effect of limitation of machine learning algorithms. Manual and automatic learning procedures are proposed for conceiving strategies for using the classifiers. Experimental results, done on a database collected from a deployed Customer Service in real conditions with real customers, show the benefits of the proposed strategies.