Mining Details from Single Opinion

In the contemporary times, different companies summarize product reviews to improve the quality of their products. Not only the companies benefit from these reviews, but these reviews also influence to a great extent the decision of the buyer. Currently a lot of research is being conducted on the opinion summaries the companies use, while the end user/buyer still has to go through whole of the opinions/reviews in order to make their decision, whether to buy a product or not. A single sentence cannot cover all the possible feature/qualities of a product. And, sometimes a single opinion may not be self-explanatory. In the proposed work, the authors have given a model that can generate different ‘signifieds’ for a single opinion. Using the definitions and examples of WordNet Dictionary, the signifieds of 80% of the opinions have been generated. The generated signifieds are not only useful for the buyer, but the person writing their opinion can also select any of these signifieds to make their point as accurately and clearly as possible.