International Conference on Mechanical and Electrical Technology, 3rd, (ICMET-China 2011), Volumes 1–3
92 Text Summarization Method Based on Normalized Non-Negative Matrix Factorization
Download citation file:
- Ris (Zotero)
- Reference Manager
This paper presents a new generic text summarization method using Non-negative Matrix Factorization (NMF) to estimate sentence relevance. Proposed sentence relevance estimation is based on normalization of NMF topic space (or feature space) and further weighting of each topic using sentences representation in topic space. Required number of sentences with the highest relevance values is selected for the summary. The number of sentences is defined by the length of the demanded summary. The developed method has been experimentally verified on DUC 2002 standard dataset and it has shown the better summarization quality and performance than state of the art methods.