Swarm based text summarization

The scoring mechanism of the text features is the unique way for determining the key ideas in the text to be presented as text summary. The treating of all text features with same level of importance can be considered the main factor causing creating a summary with low quality. In this paper, we int...

全面介紹

Saved in:
書目詳細資料
Main Authors: Wahlan, Mohammed Salem, Salim, Naomie, Suanmali, Ladda
格式: Book Section
出版: Institute of Electrical and Electronics Engineers 2009
主題:
在線閱讀:http://eprints.utm.my/id/eprint/13123/
http://dx.doi.org/10.1109/IACSIT-SC.2009.61
標簽: 添加標簽
沒有標簽, 成為第一個標記此記錄!
實物特徵
總結:The scoring mechanism of the text features is the unique way for determining the key ideas in the text to be presented as text summary. The treating of all text features with same level of importance can be considered the main factor causing creating a summary with low quality. In this paper, we introduced a novel text summarization model based on swarm intelligence. The main purpose of the proposed model is for scoring the sentences, emphasizing on dealing with the text features fairly based on their importance. The weights obtained from the training of the model were used to adjust the text features scores, which could play an important role in the selection process of the most important sentences to be included in the final summary. The results show that the human summaries H1 and H2 are 49% similar to each other. The proposed model creates summaries which are 43% similar to the manually generated summaries, while the summaries produced by Ms Word summarizer are 39% similar.