EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

2077. Java Framework for Evolutionary Computation and Multiple-Criteria Decision Analysis

Invited abstract in session MB-44: Preference Learning 2, stream Multiple Criteria Decision Analysis.

Monday, 10:30-12:00
Room: 20 (building: 324)

Authors (first author is the speaker)

1. Michał Tomczyk
Institute of Computing Science, Poznan University of Technology
2. Milosz Kadzinski
Institute of Computing Science, Poznan University of Technology

Abstract

We introduce a novel Java framework for Evolutionary Computation and multiple-criteria Decision Making named JECDM. It predominantly focuses on facilitating research on a joint area of preference-driven evolutionary multi-objective optimization. From the programmer's perspective, the main features of JECDM are as follows. Firstly, we were focused on efficiency during the development to guarantee a low computational burden imposed when executing the code. Second, the code is highly object-oriented, facilitating readability, reusability, modularity, and low code redundancy. Thirdly, we reduced the use of external libraries to a bare minimum, securing the framework's self-sufficiency. From the practitioner's viewpoint, our framework provides a vast spectrum of well-organized functionalities. They are categorized into several top-level modules dedicated to evolutionary computation, multiple-criteria decision analysis, visualization, and experimental validation. The framework is available online and free to use for academic purposes. We explain how to operate it in a series of extensive tutorials provided as PDF files accompanied by code examples that are published along with the framework. These tutorials, e.g., discuss the general structurization of the framework's components, show how to implement new algorithms by oneself or demonstrate how to design experiments even of an extensive nature without much effort spent.

Keywords

Status: accepted

Back to the list of papers