Topic-based Multi-document Summarization using Differential Evolution forCombinatorial Optimization of Sentences
Year: 2014
Type: conference-paper
Abstract: This paper describes a method of multidocument summarization with evolutionary computation.In automatic document summarization, the method to make a summary by finding the best combination of important sentences in target documents is popular approach.To find the best combination of sentences, explicit solution techniques such as integer linear programming, branch and bound method, and so on are usually adopted.However, there is a problem with them in terms of calculation efficiency.So, we apply evolutionary computation, especially differential evolution which is regarded as a method having a good feature in terms of calculation cost to obtain a reasonable quasi-optimum solution in real time, to the problem of combinatorial optimization of important sentences.Moreover, we consider latent topics in deciding the importance of a sentence, and define three fitness functions to compare the results.As a result, we have confirmed that our proposed methods reduced the calculation time necessary to make a summary considerably, although precision is more worse than the method with an explicit solution technique.
Authors: Haruka Shigematsu, Ichiro Kobayashi
Institution: Ochanomizu University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2251194501