Rhetorical Sentence Classification for Automatic Title Generation in Scientific Article

Abstract In this paper, we proposed a work on rhetorical corpus construction and sentence classification model experiment that specifically could be incorporated in automatic paper title generation task for scientific article. All rights reserved. 1.Introduction Motivated by flourishing availability...

Full description

Saved in:
Bibliographic Details
Published in:Telkomnika 2017-06, Vol.15 (2), p.656
Main Authors: Putra, Jan Wira Gotama, Khodra, Masayu Leylia
Format: Article
Language:English
Subjects:
Abstracts
Annotations
Classification
Communication
Data mining
Experiments
International conferences
Language
Linguistics
Literature reviews
Natural language processing
Scientific papers
Validity
Writers
Zoning
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract In this paper, we proposed a work on rhetorical corpus construction and sentence classification model experiment that specifically could be incorporated in automatic paper title generation task for scientific article. All rights reserved. 1.Introduction Motivated by flourishing availability of scientific articles over internet, keeping updated on relevant articles is important. Since article title represents a paper in a brief manner [1], a title is fundamental guide to quickly determine whether a paper is relevant to reader needs in literature review. Some recent studies showed that quality of article title influences its number of citation [3-5]. [...]it is very important to write a good title. According to our observation, title usually contains research aim and method [2-5], [13]. [...]Vector Space Model and Topics Model" phrase is the method. [...]useful information in this case are sentences which yield information of research purpose and method. Based on previous research, we assumed that abstract part of article is sufficient as input for title generation task, as it depicts the most important things in the paper [22, 23]. [...]we focused on extracting AIM and OWN_MTHD information in abstract part of scientific article to satisfy title communication goals. The main contributions of our research work are two-fold: (1) developing a corpus for automatic title generation task based on several research paper collections that has been annotated into one of the three categories, and (2) providing model for rhetorical sentence classification that could be incorporated in automatic title generation task. [...]full paper contains more information type...
ISSN:1693-6930
2302-9293
DOI:10.12928/TELKOMNIKA.v15İ2.4061