A novel approach to implicitly defined aspect-oriented and ontology-aided opinion mining

Organizations that sell e-commerce products over the Internet often ask their customers to rate their products/services. The number of customer evaluations is rapidly growing thanks to e-commerce, which is a growing structure. As the number of e-commerce organizations and the number of customers increases considerably, it becomes very difficult for potential customers to read these evaluations in decisionmaking about the product. At the same time, it is almost impossible for the product owner to follow these evaluations. From customer comments, extracting properties of products is an important sub-research area in the field of opinion mining (opinion mining). The extracted features help to evaluate the opinions written by customers who buy certain products. It is possible to reveal the thoughts of customers about their positive/negative experiences with these properties. In order to be able to do this, it is often necessary to develop ways of extracting product features that are explicitly and implicitly stated in reviews, since customer comments are often in the form of free text format. In this research, we aim to develop a methodology that examines the opinions/thoughts/interpretations of the products made in the Turkish language, extracts the properties of the products and summarizes the emotional activities related to these properties. In our study, the ability to identify product features expressed using synonyms or word groups can be used to identify features with greater accuracy using ontologies that include product specifications. This ability can also be used to extract implicit/explicit features that are not explicitly described in the interpretations, and to deliver successful results in extracting product attributes. Experimental evaluation of our methodology shows that our methodology gives positive results. Keywords: Text mining, opinion classification, opinion mining, ontogoly