No abstract

The spread of Internet sources has increased the volume of big data that is difficult to handle in traditional ways. So, most users need modern search systems to facilitate the search and retrieval of information in the presence of big data. However, the main challenge in the first and second conventional generations of search engines are linking different web data based on the syntax of keywords not on the semantic meaning and without a knowledge base. This manuscript proposes a framework based on modern technologies such as ETI processes, ontology graphs, and indexing RDF using wide column NoSQL technique. The main contribution of our work is introducing a mathematical model that is used to calculate the similarity score between a query and stored RDF documents based on semantic relations. Various operations were carried out to measure the proposed model's efficiency using data sources such as DBpedia, YAGO dataset. According to experimental results, the proposed model is achieving high precision compared to other related systems.