|
ИСТИНА |
Войти в систему Регистрация |
Интеллектуальная Система Тематического Исследования НАукометрических данных |
||
Text Summarization Method Based on Normalized Non-Negative Matrix Factorizationстатья
- Авторы: Tsarev D.V., Mashechkev I.V., Petrovskiy M.I.
- Сборник: Proceedings of the 3rd International Conference on Mechanical and Electrical Technology (ICMET 2011)
- Том: 1
- Год издания: 2011
- Место издания: ASME (www.asme.org) 3 Park Avenue, New York, NY 10016, USA
- Первая страница: 563
- Последняя страница: 567
- Аннотация: This paper presents a new generic text summarization method using NonnegativeMatrix Factorization (NMF) to estimate sentence relevance. Proposedsentence relevance estimation is based on normalization of NMF topic space (orfeature space) and further weighting of each topic using sentences representationin topic space. Required number of sentences with the highest relevance valuesis selected for the summary. The number of sentences is defined by the length ofthe demanded summary. The developed method has been experimentallyverified on DUC 2002 standard dataset and it has shown the bettersummarization quality and performance than state of the art methods.
- Добавил в систему: Царёв Дмитрий Владимирович