Publications

Abstract : Text summarization is essential in this fast-growing world to read the information because a vast amount of information holds various definitions among related contents. Due to this, reading loads of information documents becomes more tedious. Most text summarization techniques are based on information extraction from unstructured documents, leading to more non-residual abstraction in sentence case analysis. To resolve this problem, a Lattice abstraction-based content summarization (Labs-CS) is proposed to reduce the unstructured documents using the Intra sub-cluster to precipitate sentences. Initially, this proposed method preprocesses natural language processing with a dictionary of terms to make corpus reader content analysis and then de-noises the contents by eliminating the nonstructural text in segmented sentences. Depending on the structural segmentation, the key terms are grouped into clusters and summarized in the sentences into intra-cluster comparisons in another cluster. It creates a lattice-based essential term fragmentation; the text terms are splatted into residual and non-residual terms, then the residual terms are compared with a dictionary of syntactic words which are extracted. Based on the extracted terms, Baseline Abstractive Sentences (BAS) are created using Lexical Chaining Progress (LCP). Finally, the syntactic sequence analyzer combines the extracted term to summarize a document. The proposed system produces high performance by achieving high coherence to reduce the complexity of summarized multilingual documents.