id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_nlcqi2h725a3vh67xurzf5wy7a	Mahmood Yousefi-Azar	Text summarization using unsupervised deep learning	2017	15	.pdf	application/pdf	10671	949	65	We present methods of extractive query-oriented single-document summarization using a deep auto-encoder (AE) to investigate the effect of adding small random noise to local tf as the input representation of AE, and propose an ensemble of In our case, after ranking the sentences of a document based on the cosine similarity, they must be selected to generate algorithms, in particular, Restricted Boltzmann Machine (RBM) and AEs. The word representation is presented in section Table 1: ROUGE-2 recall of subject-oriented summarization for tf-idf, tf and models versus the number of selected sentences. Table 2: ROUGE-2 recall of key-phrase-oriented summarization for tf-idf, tf and models versus the number of selected sentences. Table 4: ROUGE-2 recall of subject-oriented summarization for tf-idf, Ltf and models versus the number of selected sentences. In this paper, we presented a query-based single-document summarization scheme using an unsupervised deep neural network.	./cache/work_nlcqi2h725a3vh67xurzf5wy7a.pdf	./txt/work_nlcqi2h725a3vh67xurzf5wy7a.txt