Summarization holds vital significance in the age of a data-centric world. This research paper tries to retain an overall summary of the article using an optimal multi-document integration. Nevertheless, location-based extractions are biased, and the data loss is high compared to the proposed approach that derives a matrix subset from a random permutation of elements after which extraction takes place. The proposed extractive method of integrating multi-documents has two stages. Initially, an iterative elimination of matrix subsets is done, finally the documents are integrated, and their lexical similarity is computed. The retention rates recorded in this implementation are considerably high even when the compression rates are increased. As proof of concept, our experiment results of extractive integration reveal high retention rate that could improve the quality of the generated summaries.
CITATION STYLE
George, A., & Hanumanthappa, M. (2019). Optimal Multi-document Integration Using Iterative Elimination and Cosine Similarity. In Advances in Intelligent Systems and Computing (Vol. 841, pp. 699–705). Springer Verlag. https://doi.org/10.1007/978-981-13-2285-3_82
Mendeley helps you to discover research relevant for your work.