Fine Analysis of opinions in corpuses of Customer Satisfaction

Starting: 01/10/2019
PhD Student: Rémi Bouvet
Advisor(s): Sylvain Meignier (LIUM - LST)
Co-advisor(s): Nathalie Camelin (LIUM - LST) , Nicolas Dugué (LIUM - LST, IEIAH)
Funding: Contrat de collaboration MMA

This thesis is part of a scientific collaboration between LIUM and MMA on the fine analysis of opinions. MMA, the leading player in property and casualty insurance in France, is very committed to analysing customer paths. In a customer relationship strategy, MMA wants to determine the factors of satisfaction, commitment and recommendation. Its marketing department distributes satisfaction surveys to collect numerous textual feedback from customers. The volume collected is such that it is impossible to manually process all of this data. The MMA group called upon LIUM’s knowledge to automatically analyze opinion information. Thus, the purpose of the thesis is to propose an experimental prototype to obtain a detailed and relevant analysis of the subjective content of MMA corpuses. In this context, several scientific issues will be addressed, including: detection of subjectivity, construction of a robust and relevant continuous representation space, fine detection of opinion by considering its target, its polarity and its intensity.