Shuffled Frog-Leaping Algorithm Metaheuristic for Extractive Single- Document Summarization
Due to the increasing amount of information available on the Internet, it is important for users to have a summary containing the most important ideas from the documents they find, in order to quickly identify which ones to read. This article addresses this issue through a modified algorithm for th...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universidad Distrital Francisco José de Caldas
2024-12-01
|
| Series: | Revista Científica |
| Subjects: | |
| Online Access: | https://revistas.udistrital.edu.co/index.php/revcie/article/view/22578 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Due to the increasing amount of information available on the Internet, it is important for users to have a summary containing the most important ideas from the documents they find, in order to quickly identify which ones to read. This article addresses this issue through a modified algorithm for the automatic generation of single-document extractive summaries, aiming to produce summaries of a quality comparable to those generated by expert humans. This proposal is based on the shuffled frog-leaping metaheuristic algorithm (SFLA) and includes a global explicit tabu memory. Its goal is to optimize a weighted objective function with characteristics such as length (measured in words), position within the document, similarity to the document's title, cohesion (similarity between the sentences in the summary), and coverage (similarity between the sentences in the summary and the document). To this effect, an iterative research procedure was followed, consisting of four stages (observation, problem identification, development, and solution testing) over two iterative cycles. In the first cycle, the initialization and evolution schemes were analyzed and selected to modify the base algorithm. This, in addition to parameter tuning. In the second cycle, a tabu memory was selected for integration into the proposed algorithm, and the corresponding tuning was performed. To evaluate the quality of the summaries generated by our proposal, ROUGE metrics were used on the DUC datasets. The results are comparable to and surpass those of various methods in the state of th art. The proposed algorithm stands out for its simplicity of implementation and the reduced number of objective function evaluations, which implies lower computation times.
|
|---|---|
| ISSN: | 0124-2253 2344-8350 |