‘Long autonomy or long delay?’ The importance of domain in opinion mining

This work tackles the task of opinion extraction in a domain-oriented approach, defining a set of domain-specific resources which capture valuable knowledge about how people express opinions on a given domain and automatically induced from an set of annotated documents.

Abstract

Nowadays, people do not only navigate the web, but they also contribute contents to the Internet. Among\n\t\t\t\t other things, they write their thoughts and opinions in review sites, forums, social networks, blogs and\n\t\t\t\t other websites. These opinions constitute a valuable resource for businesses, governments and consumers.\n\t\t\t\t In the last years, some researchers have proposed opinion extraction systems, mostly domain-independent\n\t\t\t\t ones, to automatically extract structured representations of opinions contained in those texts. In\n\t\t\t\t this work, we tackle this task in a domain-oriented approach, defining a set of domain-specific resources\n\t\t\t\t which capture valuable knowledge about how people express opinions on a given domain. These\n\t\t\t\t resources are automatically induced from a set of annotated documents. Some experiments were carried\n\t\t\t\t out on three different domains (user-generated reviews of headphones, hotels and cars), comparing our\n\t\t\t\t approach to other state-of-the-art, domain-independent techniques. The results confirm the importance\n\t\t\t\t of the domain in order to build accurate opinion extraction systems. Some experiments on the influence\n\t\t\t\t of the dataset size and an example of aggregation and visualization of the extracted opinions are also\n\t\t\t\t shown.