| ITI əməkdaşlarının elmi işləri - məqalə |
| Biblioqrafik təsvir | |
| Alguliyev , R.M. An unsupervised approach to generating generic summaries of documents / R.M. Alguliyev , R.M. Aliguliyev , N. Isazade // Applied Soft Computing. - 2015. - N: 34, vol.13.- P. 236-250. | |
| Annotasiya | |
| We present an optimization-based unsupervised approach to automatic document summarization. In theproposed approach, text summarization is modeled as a Boolean programming problem. This model gen-erally attempts to optimize three properties, namely, (1) relevance: summary should contain informativetextual units that are relevant to the user; (2) redundancy: summaries should not contain multiple textualunits that convey the same information; and (3) length: summary is bounded in length. The approachproposed in this paper is applicable to both tasks: single- and multi-document summarization. In bothtasks, documents are split into sentences in preprocessing. We select some salient sentences from docu-ment(s) to generate a summary. Finally, the summary is generated by threading all the selected sentencesin the order that they appear in the original document(s). We implemented our model on multi-documentsummarization task. When comparing our methods to several existing summarization methods on anopen DUC2005 and DUC2007 data sets, we found that our method improves the summarization resultssignificantly. This is because, first, when extracting summary sentences, this method not only focuses onthe relevance scores of sentences to the whole sentence collection, but also the topic representative ofsentences. Second, when generating a summary, this method also deals with the problem of repetitionof information. The methods were evaluated using ROUGE-1, ROUGE-2 and ROUGE-SU4 metrics. In thispaper, we also demonstrate that the summarization result depends on the similarity measure. Resultsof the experiment showed that combination of symmetric and asymmetric similarity measures yieldsbetter result than their use separately. | |
| Elektron variant | |
| Elektron variant |
________
© ict.az http://ict.az/az