A recursive framework for improving the performance of multi-objective differential evolution algorithms for gene selection

Swarm and Evolutionary Computation, Год журнала: 2024, Номер 87, С. 101546 - 101546

Опубликована: Апрель 4, 2024

Язык: Английский

---

Evolving cybersecurity frontiers: A comprehensive survey on concept drift and feature dynamics aware machine and deep learning in intrusion detection systems

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 137, С. 109143 - 109143

Опубликована: Авг. 22, 2024

Язык: Английский

---

A high-dimensional feature selection algorithm via fast dimensionality reduction and multi-objective differential evolution

Swarm and Evolutionary Computation, Год журнала: 2025, Номер 94, С. 101899 - 101899

Опубликована: Март 4, 2025

Язык: Английский

---

Robust sparse and low-redundancy multi-label feature selection with dynamic local and global structure preservation

Pattern Recognition, Год журнала: 2022, Номер 134, С. 109120 - 109120

Опубликована: Окт. 23, 2022

Язык: Английский

---

Location and path planning for urban emergency rescue by a hybrid clustering and ant colony algorithm approach

Applied Soft Computing, Год журнала: 2023, Номер 147, С. 110783 - 110783

Опубликована: Авг. 26, 2023

Язык: Английский

---

Feature subset selection in structural health monitoring data using an advanced binary slime mould algorithm

Journal of Structural Integrity and Maintenance, Год журнала: 2023, Номер 8(4), С. 209 - 225

Опубликована: Июль 6, 2023

Feature Selection (FS) is an important step in data-driven structural health monitoring approaches. In this paper, Advanced version of the Binary Slime Mould Algorithm (ABSMA) introduced for feature subset selection to improve performance damage classification techniques. Two operators mutation and crossover are embedded algorithm, overcome stagnation situation involved (BSMA). The proposed ABSMA then a new SHM framework which consists three main steps. first step, time domain responses collected pre-processed extract statistical features. second order extracted features reduced using optimization algorithm find minimal salient by removing irrelevant, redundant data. Finally, optimized vectors used as inputs Neural Network (NN) based models. Benchmark datasets timber bridge model three-story frame structure employed validate algorithm. results show that provides better convergence rate compared other commonly binary algorithms.

Язык: Английский

---

Dynamic feature selection model for adaptive cross site scripting attack detection using developed multi-agent deep Q learning model

Journal of King Saud University - Computer and Information Sciences, Год журнала: 2023, Номер 35(6), С. 101490 - 101490

Опубликована: Янв. 23, 2023

Web applications' popularity has raised attention in various service domains, which increased the concern about cyber-attacks. One of these most serious and frequent web application attacks is a Cross-site scripting attack (XSS). It causes grievous harm to victims. Existing security methods against XSS fail due evolving nature attacks. aspect feature drift changes relevancy degradation performance. Unfortunately, dynamic awareness occurrence missing. Thus, this study attempts fill gap by proposing drift-aware algorithm for detecting evolved The proposed approach selection based on deep Q-network multi-agent (DQN-MAFS) framework. Each agent associated with one responsible selecting or deselecting its feature. DQN-MAFS provides sub-model reward distribution over agents, named as fair FARD-DFS. This framework capable supporting real-time, updates adjustment embedded knowledge long new labelled data arrives. been evaluated using four real datasets length sizes. evaluation process was conducted compared state-of-the-art works. obtained results show superiority our FARD-DFS benchmarks terms majority metrics. improvement percentages mean accuracy F1-measure ranged from 1.01% 12.1% 0.55% 6.88%, respectively, comparison benchmarks. can be deployed an autonomous detection system without need any offline retraining model detect attack.

Язык: Английский

---

Multi-objective binary grey wolf optimization for feature selection based on guided mutation strategy

Applied Soft Computing, Год журнала: 2023, Номер 145, С. 110558 - 110558

Опубликована: Июнь 19, 2023

Язык: Английский

---

Minimizing the searching time of multiple targets in uncertain environments with multiple UAVs

Applied Soft Computing, Год журнала: 2024, Номер 155, С. 111471 - 111471

Опубликована: Март 6, 2024

The focus of this paper is the use Unmanned Aerial Vehicles (UAVs) for searching multiple targets under uncertain conditions in minimal possible time. problem, known as Minimum Time Search (MTS), belongs to Probabilistic (PS) field and addresses critical missions, such search & rescue, military surveillance. These operations, characterized by complex environments, demand efficient UAV trajectory optimization. multi-target version PS introduces additional challenges, due their higher complexity need wisely distribute UAV's efforts among targets. In order tackle under-explored aspect MTS, we optimize time find all with new Ant Colony Optimization (ACO)-based planner. This novel optimization criterion formulated using Bayes' theory, considering probability models (initial belief motion model) sensor likelihood. Our work contributes significantly (i) developing an objective function tailored (ii) proposing ACO-based planner designed effectively handle complexities moving targets, (iii) introducing a constructive heuristic that used planner, specifically MTS problem. efficacy our approach demonstrated through comprehensive analysis validation across various scenarios, showing superior performance over existing methods problems.

Язык: Английский

---

LSFSR: Local label correlation-based sparse multilabel feature selection with feature redundancy

Information Sciences, Год журнала: 2024, Номер 667, С. 120501 - 120501

Опубликована: Март 20, 2024

Язык: Английский

---