Aspect term extraction and optimized deep fuzzy clustering-based inverted indexing for document retrieval

Abstract

Finding good relevant documents for query optimization is a well-known difficulty in the field of document retrieval. This paper develops a novel approach, named Exponential Aquila Optimizer (EAO)-based Deep Fuzzy Clustering for retrieving the documents. The proposed technique effectively finds the relevant documents and tries to understand the relationship among the documents and queries in terms of the significance of documents for query optimization. Here, the Deep Fuzzy Clustering is employed for performing cluster-based inverted indexing where the Training procedure of Deep Fuzzy Clustering is done using the developed optimization algorithm, named EAO. Meanwhile, the developed EAO is newly designed by the incorporation of EWMA and AO. In addition, complex query matching is done using the Tversky index for the user-based queries, such as multigram queries and semantic queries. On the other hand, the RV coefficient is accomplished for performing query optimization for relevant document retrieval. The proposed technique achieves better performance in terms of the performance metrics, like precision, recall, and F-measure with the maximum precision of 1, maximum recall of 0.956, and maximum F-measure of 0.977, respectively.