Ontology-Based Product´s Reputation Model

Different reputation models are used in the web in order to generate reputation values for products using uses´ review data. Most of the current reputation models use review ratings and neglect users´ textual reviews, because it is more difficult to process. However, we argue that the overall reputation score for an item does not reflect the actual reputation for all of its features. And that´s why the use of users´ textual reviews is necessary. In our work we introduce a new reputation model that defines a new aggregation method for users´ extracted opinions about products´ features from users´ text. Our model uses features ontology in order to define general features and sub-features of a product. It also reflects the frequencies of positive and negative opinions. We provide a case study to show how our results compare with other reputation models.