Deep Learning-based Information Extraction and Retrieval for Web Semantic

Abstract

The semantic web is the requirement of the current trend of web data engineering. In web data engineering, the extraction of information and the minimization of feature points of extraction move to the next generation of web-based data retrieval. The ontology-based information extraction reduces the gap in information retrieval. In this paper proposed deep learning-based information extraction and retrieval in the domain ontology. The deep learning enriches the process of information extraction with ontology. The extracted feature points from the deep learning minimized some features of extracted information using firefly algorithms. The firefly algorithm is bioinspired algorithms used for the process of optimization, and the nature of the algorithm is dynamic and iterative. The proposed algorithm is simulated in MATLAB environments and used thee ontology information models such as education, news, and medical. For the extraction of information used 200 web pages data relevant to this ontology. The proposed algorithm compares with SVM and other algorithms and measures empirical results such as precision, recall, and accuracy. The analysis of results suggests that the proposed algorithm is better than the support vector machine.