Sentiment Analysis: What's Your Opinion?

For more than 10 years now, Sentiment Analysis has enjoyed enormous popularity in Computational Linguistics, one main reason being its great potential for practical applications, predominantly (but not only) for industrial purposes. We observe a tendency that early work referred to certain theoretical notions of Subjectivity, whereas a lot of the later approaches follow the ‘engineering’ perspective that can include using terminology somewhat indiscriminately and are not aiming at making progress with the underlying theoretical issues. In this paper, we first survey some important notions surrounding “Subjectivity” in Linguistics and Psychology, trying to broaden the perspective of standard opinion analysis. Thereafter, we take a snapshot of the state of the art in computational Sentiment Analysis, as it has developed since roughly 2000. Combining these two viewpoints leads us to assessing the gap between the broader notion of Subjectivity Analysis and the subfields that language technology research tends to focus on. We suggest a few potential research directions that could help narrowing this gap.