Towards the Extraction of Customer-to-Customer Suggestions from Reviews

State of the art in opinion mining mainly focuses on positive and negative sentiment summarisation of online customer reviews. We observe that reviewers tend to provide advice, recommendations and tips to the fellow customers on a variety of points of interest. In this work, we target the automatic detection of suggestion expressing sentences in customer reviews. This is a novel problem, and therefore to begin with, requires a well formed problem definition and benchmark dataset. This work provides a 3-fold contribution, namely, problem definition, benchmark dataset, and an approach for detection of suggestions for the customers. The problem is framed as a sentence classification problem, and a set of linguistically motivated features are proposed. Analysis of the nature of suggestions, and classification errors, highlight challenges and research opportunities associated with this problem.