求翻译：Automatic generic document summarization based on unsupervised schemes is very useful approaches because it does not require training data. Although techniques using latent semantic analysis (LSA) and non-negative matrix factorization (NMF) have been applied to determine topics of documents, there are no researches on reduction of matrix and speeding up of computation of the NMF method. In order to achieve this scheme, this paper utilizes the generic impressive expressions from newspapers to extract important sentences as summary. Therefore, it has no stemming processes and no filtering of stop words. Generally, novels are typical documents providing sentimental impression for readers. But, newspapers deliver different impressions for new knowledge because it informs readers of new information about current events, informative articles and diverse features. The proposed method introduces impressive expressions for newspapers and their measurements are applied to the NMF method. Form 100 KB text data of experimental results by the proposed method, it turns out that the matrix size reduce by 80% smaller and the speeding up of the computation of the NMF method becomes 7 times faster than original method, without degrading the relevancy of extracted sentences.是什么意思？

待解决 悬赏分：1 - 离问题结束还有