Extracting knowledge from various sources is a tedious task. Multi-document Text Summarization (MDTS) aims to extrapolate data from various sources and present it in a concise, cohesive form, in a way that is simple for the reader to comprehend and yet ensures that they obtain the important information. A meta-heuristic optimization algorithm called the firefly algorithm is utilized to generate a summary. Topic Relevance Factor, Coherency Factor, and Readability Factor are used to establish the fitness function. We use BERT-based similarity to calculate these factors which are then later input to the fitness function. The experiments are conducted on DUC 2003 and DUC 2004 datasets. The suggested algorithm’s performance is compared to that of previous meta-heuristic and graph-based techniques.
Rushi Desai, Saloni Patel, Shourya Kothari, Pankaj Sonawane
Dwarkadas Jivanlal Sanghvi College of Engineering, India
Multiple document summarization, Meta-heuristic optimization, Firefly algorithm, ROUGE, BERTscore
| January | February | March | April | May | June | July | August | September | October | November | December |
| 1 | 1 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Published By :
ICTACT
Published In :
ICTACT Journal on Data Science and Machine Learning
( Volume: 4 , Issue: 4 , Pages: 520 - 525 )
Date of Publication :
September 2023
Page Views :
256
Full Text Views :
15
|