Sentiment analysis is used to quantify the sentiment of unstructured text, which could be used to draw insights on public opinions about products, services, or any topic that is of relevance. Most of the available sentiment analysis implementations provide an overall sentiment which is not sufficient because it does not provide an aspect-based sentiment. In this work, we present an extendible ontology-based approach for aspect-based sentiment analysis. In this work, we propose a novel and extendible Ontological approach for aspect-based Sentiment Analysis. It also focuses on learning the aspects using topic modelling to automate the aspect detection. The results indicate that the proposed approach can be effective in classifying sentiments under different aspects. The proposed approach may have applications in domains such as e-commerce web sites, product reviewing platforms and social media platforms.