Sentiment Analysis (SA) refers to the use of Natural Language Processing to identify subjective information from textual material. It aims at determining the attitude of a speaker or a writer with respect to some topic or the overall contextual polarity of a document. It is even more important in the World Wide Web era, because web 2.0 and social networks technologies allow people to generate contents on blogs, forums, writing their opinion about facts, things, events. Collective knowledge has spread throughout the Web, particularly in areas related to everyday life, such as commerce, tourism, education, and health. The interest in the analysis of what people write, and in particular on the way they express is rapidly growing.
Sentiment Analysis is generally seen in three different modalities:
- Sentiment Classification. It considers opinion mining as a text classification problem: a text is associated to classes reflecting their positive or negative attitude. For example, given a product review, the system should determine whether the review expresses a positive or a negative sentiment.
- Aspect-Based Opinion Mining. This modality consider the word level to discover details about the sentiment expressed. For example, in a product review, this task first identifies product features, and then classify each of them with respect to sentiment.
- Comparative Sentence. Comparison is another type of evaluation, which directly compares one object against one or more other similar objects.
SAG group works on Sentiment Analysis exploiting data-driven approaches on different texts, such as blogs or twitter material.
Bing Liu. Sentiment Analysis and Opinion Mining. Morgan & Claypool Publishers, May 2012
Bo Pang and Lillian Lee. Foundations and Trends in Information Retrieval 2(1-2), pp. 1–135, 2008.
Minqing Hu , Bing Liu. Mining opinion features in customer reviews, Proceedings of the 19th national conference on Artifical intelligence, p.755-760, July 25-29, 2004, San Jose, California