Multi-Document Summarization Based on Two-Level Sparse Representation Model

• DOI: 10.1609/aaai.v29i1.9161
• Publication Date: 2022-06-23T23:20:00Z
• Authors: He Liu; Hongliang Yu; Zhi-Hong Deng
• Keywords: Multi-document summarization; Neural coding; Benchmark (surveying); Representation

Multi-document summarization is of great value to many real world applications since it can help people get the main ideas within a short time.In this paper, we tackle problem extracting summary sentences from multi-document sets by applying sparse coding techniques and present novel framework challenging problem. Based on data reconstruction sentence denoising assumption, two-level representation model depict process summarization. Three requisite properties proposed form an ideal reconstructable summary: Coverage, Sparsity Diversity. We then formalize task as optimization according above properties, use simulated annealing algorithm solve it.Extensive experiments benchmark DUC2006 DUC2007 show that our effective outperforms state-of-the-art algorithms.