Paper Info
Title
Korean Document Summarization Using Topic Phrases Extraction and Locality-Based Similarity
Abstract
We describe a hybrid approach to summarize the document content using the topic phrases extraction and the query-based summary. The topic phrases are extracted using machine learning algorithm. We use these topic phrases as the query terms with locality-based similarity calculation in order to extract highly ranked sentences or paragraph. We experiment with three machine learning methods, Naive Bayesian, decision tree and supported vector machine, for extracting the topic phrases effectively and discuss the results. The overall summaries have been evaluated for the extraction accuracy compared with the human-selected summaries.
Year
DOI
Venue
2003
10.1007/978-3-540-39592-8_44
Lecture Notes in Computer Science
Keywords
Field
DocType
decision tree,support vector machine,machine learning
Decision tree,Locality,Ranking,Naive Bayes classifier,Computer science,Support vector machine,Phrase,Paragraph,Natural language processing,Artificial intelligence,Sentence,Machine learning
Conference
Volume
ISSN
Citations 
2871
0302-9743
0
PageRank 
References 
Authors
0.34
8
3
Name
Order
Citations
PageRank
Je Ryu101.01
Kwang-Rok Han221.41
Kee-Wook Rim315424.20