Metrik

  • visibility 64 kali dilihat
  • get_app 20 downloads
description Journal article public Journal of Intelligent Systems

Hybrid Keyword Extraction Algorithm and Cosine Similarity for Improving Sentences Cohesion in Text Summarization

Rizki Darmawan, Romi Satria Wahono
Diterbitkan 2015

Abstrak

As the amount of online information increases, systems that can automatically summarize text in a document become increasingly desirable. The main goal of a text summarization is to present the main ideas in a document in less space. In the create text summarization, there are two procedures which are extraction and abstraction procedure. One of extraction procedure is using keyword extraction algorithm which is easier and common but has problems in the lack of cohesion or correlation between sentences. The cohesion between sentences can be applied by using a cosine similarity method. In this study, a hybrid keyword extraction algorithm and cosine similarity for improving sentences cohesion in text summarization has been proposed. The proposed method using compression various compression ratios is used to create candidate of the summary. The result show that proposed method could affect significant increasing cohesion degree after evaluated in the t-Test. The result also shows that 50% compression ratio obtains the best result with Recall, Precision, and F-Measure are 0.761, 0.43 and 0.54 respectively; since summary with compression ratio 50% has higher intersection with human summary than another compression ratio.

Full text

 

Metrik

  • visibility 64 kali dilihat
  • get_app 20 downloads