A Comprehensive Survey of Scalable Multi-Document Summarization using Natural Language Processing

Abstract

Natural Language Processing (NLP) approaches are the major objectives for users in this emergence of the Internet. Researchers have recently been interested in multi-document summarization (MDS) because of the difficulties it provides in producing well-summarized results. MDS is also an effective method for aggregating information since it creates a comprehensive and brief summary from a collection of topic-related papers. Document size limitations, limited memory in computer resources, and redundancy of identical words in numerous documents are all difficulties that MDS has to deal with. Natural Language Processing (NLP) methods are used in this paper to address these difficulties in MDS. Our study is the first of its type and provides a comprehensive overview of current NLP-based multi-document summarization methods via a suggested taxonomy. In addition, this paper offers a new approach for summarizing NLP design techniques and performing a systematic state-of-the-art review. The differences between various NLP approaches are highlighted, that are typically addressed in the existing literature. Furthermore, we discuss some future directions for this field's new and innovative advancement.

Keywords: Natural Language Processing; Multi-Document Summarization; Scalable; Data Extraction; Text Summarization.