Single document text summarization technique using optimal combination of cuckoo search algorithm, sentence scoring and sentiment score

Data mining or Knowledge Discovery Database (KDD) is a process of digging through the huge volume of data for finding out the hidden pattern and rules. Text summarization is a technique of data mining which is used to represent text document in a concise manner. Text summarization methods are classi...

Full description

Saved in:
Bibliographic Details
Published in:International journal of information technology (Singapore. Online) Vol. 13; no. 5; pp. 1805 - 1813
Main Authors: Mandal, Shrabanti, Singh, Girish Kumar, Pal, Anita
Format: Journal Article
Language:English
Published: Singapore Springer Singapore 01.10.2021
Springer Nature B.V
Subjects:
Algorithms
Artificial Intelligence
Computer Imaging
Computer Science
Data mining
Image Processing and Computer Vision
Machine Learning
Original Research
Pattern Recognition and Graphics
Search algorithms
Sentiment analysis
Software Engineering
Vision
Content-based metric
Sentiment score
Sentence scoring method
Cuckoo Search algorithm
ISSN:2511-2104, 2511-2112
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data mining or Knowledge Discovery Database (KDD) is a process of digging through the huge volume of data for finding out the hidden pattern and rules. Text summarization is a technique of data mining which is used to represent text document in a concise manner. Text summarization methods are classified as abstractive, extractive, indicative, informative, single document and multiple documents. In single document approach only one document is summarized and in multi-document summarization multiple documents are summarized. In abstractive approach document(s) is summarized using newly composed sentences while in extractive summarization existing sentences from the document(s) is used to summarize document. Summary size indicates the summarization system as indicative or informative. This research work proposes a method for single document extractive summarization. The proposed method is based on sentence scoring, Cuckoo Search (CS) algorithm and sentiment analysis. Sentence scoring methods are used to represent sentences into numerical forms and then CS algorithm is used to select the best suitable sentences that can be used to give the extractive summary. Sentiment analysis is used to select most significant sentences to represent the summary. Experimental result shows that proposed method produces notable improvement in terms of precision, recall and F1-score to existing three methods namely CSSA, summary using key concepts and sentence importance and DUC baseline.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2511-2104
2511-2112
DOI:10.1007/s41870-021-00739-2