Opinion target extraction is a crucial task of opinion mining, aiming to extract occurrences of the different entities of a corpus that are subjects of an opinion. In order to produce a readable and comprehen-sible opinion summary, these occurrences are aggregated under higher order labels, or entities, in a second task. In this paper we argue that combining the two tasks, i.e. extracting opinion targets using entities as labels instead of binary labels, yields better results for opinion extraction. We compare the binary and the multi-class approaches on available datasets in English and French, and conduct several investigation experiments to explain the promising results. Our experiments show that an entity-based labelling not only improves opinion extraction in a single domain setting, but also let us combine training data from different domains to improve the extraction, a result that has never been observed on target-based training data.