Scalable Ontology-Driven Data Mining Algorithms for Real-Time Analysis of IoT Data Streams
Ala Harika, K Aravinda, Anurag Shrivastava, Amandeep Nagpal, Praveen Praveen, Sadeq Khudhur Thajeel · 2024
The emergence of Internet of Things (IoT) technology has resulted in an unparalleled surge of data streams, requiring sophisticated approaches for instantaneous analysis. This research study presents scalable ontology-driven data mining techniques that aim to improve the efficiency and accuracy of extracting information from real-time IoT data streams. The suggested system combines semantic web technologies with data mining to tackle the inherent issues of velocity, diversity, and volume of IoT data. The technique guarantees dynamic scalability and enhanced contextual comprehension by combining ontology-based data architecture with adaptive data mining technologies. The ontology functions as a semantic layer, enabling the seamless integration of dynamic data, the representation of knowledge, and the processing of queries. Meanwhile, the adaptive algorithms aim to optimize performance in real-time and ensure efficient utilization of resources. The research assesses the architecture by conducting comprehensive experiments using actual IoT information, showcasing substantial improvements in decision-making velocity and data throughput. This study makes a valuable contribution to the area by providing a strong and effective solution for analyzing real-time data from the IoT. This solution allows for intelligent decision-making in several domains, including smart cities, healthcare, and industrial automation. The use of an ontology-driven approach not only improves the understanding of data and increases the speed at which it is processed, but also allows for adaptation to changing data streams. This ensures consistent performance even in the face of a constant flood of data from the IoT.