Understanding opinionated texts containing people’s opinion and sentiment can be problematic owing to the large volume of consumer opinion that exists in free text. Natural language processing is a difficult subfield of artificial intelligence and significant hand-labelled data is required for high-quality sentiment analysis owing to the fine-grained nature of consumer reviews that could have different meanings. For example, “the food was great but the service was disappointing”
Aspect-based Sentiment Analysis System using state-of-the-art artificial intelligence methods to discover sentiment towards different aspects of products/services in free text and sentiment analysis system for finding fine-grained opinions in consumer reviews. A new hand-crafted data set labelled with sentiment expressions and clues The design of state-of-the-art machine learning models for high-quality sentiment analysis Transferable technology to more general opinion mining from the text (e.g. opinions about people, legislations, etc.)
The ADAPT Centre for Digital Content Technology is funded under the SFI Research Centres Programme (Grant 13/RC/2106) and is co-funded under the European Regional Development Fund. Rasoul Kaljahi and Jennifer Foster
