Research Repository

ECoFFeS: A Software Using Evolutionary Computation for Feature Selection in Drug Discovery

Liu, ZZ and Huang, JW and Wang, Y and Cao, DS (2018) 'ECoFFeS: A Software Using Evolutionary Computation for Feature Selection in Drug Discovery.' IEEE Access, 6. 20950 - 20963. ISSN 2169-3536

[img]
Preview
Text
08328818.pdf - Published Version

Download (9MB) | Preview

Abstract

Feature selection is of particular importance in the field of drug discovery. Many methods have been put forward for feature selection during recent decades. Among them, evolutionary computation has gained increasing attention owing to its superior global search ability. However, there still lacks a simple and efficient software for drug developers to take advantage of evolutionary computation for feature selection. To remedy this issue, in this paper, a user-friendly and standalone software, named ECoFFeS, is developed. ECoFFeS is expected to lower the entry barrier for drug developers to deal with feature selection problems at hand by using evolutionary algorithms. To the best of our knowledge, it is the first software integrating a set of evolutionary algorithms (including two modified evolutionary algorithms proposed by the authors) with various evaluation combinations for feature selection. Specifically, ECoFFeS considers both single-objective and multi-objective evolutionary algorithms, and both regression- and classification-based models to meet different requirements. Five data sets in drug discovery are collected in ECoFFeS. In addition, to reduce the total analysis time, the parallel execution technique is incorporated into ECoFFeS. The source code of ECoFFeS can be available from https://github.com/JiaweiHuang/ECoFFeS/.

Item Type: Article
Uncontrolled Keywords: Evolutionary computation, feature selection, drug discovery, single-objective optimization, multi-objective optimization, parallel execution
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Science and Health > Computer Science and Electronic Engineering, School of
Depositing User: Elements
Date Deposited: 26 Jun 2018 13:19
Last Modified: 26 Jun 2018 14:15
URI: http://repository.essex.ac.uk/id/eprint/22304

Actions (login required)

View Item View Item