Case-Studies in Mining User-Generated Reviews for Recommendation
User-generated reviews are now plentiful online and they have proven to be a valuable source of real user opinions and real user experiences. In this chapter we consider recent work that seeks to extract topics, opinions, and sentiment from reviewtext that is unstructured and often noisy.We describe and evaluate a number of practical case-studies for howsuch information can be used in an information filtering and recommendation context, from filtering helpful reviews to recommending useful products.
National University of Ireland, Dublin (UCD)
Open access repository: