Machine Learning and Ontology based Framework in Information Retrieval using Semantic Query Expansion

Data and information stored online is steadily growing in volume. Indeed, when it comes to readily available, high-quality information, the web is unrivalled. Information retrieval describes the steps used to find the desired data in a database. By analyzing the user’s query semantically, semantic information retrieval can offer results that are more relevant to their original intent. When this occurs, ontology-based knowledge representation has the potential to facilitate semantic retrieval more efficiently than alternative representation approaches, such as frames and semantic networks. In the past twenty years, machine learning techniques have found widespread application across several fields of art, science, and technology. Information retrieval activities that make extensive use of machine learning techniques include document analysis and query expansion. This paper presented a semantic information retrieval system based on machine learning and ontologies. Our research demonstrates that the method outperforms the state-of-the-art on certain recall, precision, and f-measure metrics used to evaluate the effectiveness of semantic information retrieval. Experiments have proven that the method helps the system convey knowledge better and retrieves semantic information faster.