Loading...

Media is loading
 

Paper Type

ERF

Abstract

Titles contain rich information, making them advantageous for extracting the overall topic of a text or a set of key words. However, due to their brief nature, there is skepticism and challenge regarding their suitability for machine learning techniques that require statistical inference. This study evaluates the effectiveness of traditional topic modeling methods and TextRank in summarizing document content through titles. It investigates the delivery of effective summary meanings using intrinsic methods via the Analytical Hierarchy Process approach. Furthermore, the research examines the limitations of topic modeling and the efficiency of TextRank in handling short sentences, aiming to underline the intricacies involved in assessing summary quality and the necessity for comprehensive strategies in summarization research. As a result, we found that among the four performance evaluation factors in summarization, information significance was the most important factor. Additionally, TextRank is a more effective method for summarizing titles compared to topic modeling methods.

Paper Number

1703

Author Connect URL

https://authorconnect.aisnet.org/conferences/AMCIS2024/papers/1703

Comments

SIGDSA

Author Connect Link

Share

COinS
Best Paper Nominee badge
Top 25 Paper Badge
 
Aug 16th, 12:00 AM

Comparing Summarization Techniques for Tourism Event Articles: Topic Modeling vs. TextRank

Titles contain rich information, making them advantageous for extracting the overall topic of a text or a set of key words. However, due to their brief nature, there is skepticism and challenge regarding their suitability for machine learning techniques that require statistical inference. This study evaluates the effectiveness of traditional topic modeling methods and TextRank in summarizing document content through titles. It investigates the delivery of effective summary meanings using intrinsic methods via the Analytical Hierarchy Process approach. Furthermore, the research examines the limitations of topic modeling and the efficiency of TextRank in handling short sentences, aiming to underline the intricacies involved in assessing summary quality and the necessity for comprehensive strategies in summarization research. As a result, we found that among the four performance evaluation factors in summarization, information significance was the most important factor. Additionally, TextRank is a more effective method for summarizing titles compared to topic modeling methods.

When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.