Review
Automation & Control Systems
Ebubekir Kaya, Beyza Gorkemli, Bahriye Akay, Dervis Karaboga
Summary: The ABC algorithm is a popular optimization algorithm that has been successfully applied to solve real-world problems. This study examines combinatorial optimization approaches based on the ABC algorithm, provides summaries of related studies, and introduces the ABC algorithm-based approaches used. The study also evaluates mechanisms to improve the local search capability of the ABC algorithm and analyzes neighborhood operators, selection schemes, initial populations determination approaches, hybrid approaches, and test instances used in evaluating the performances of ABC algorithms.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Information Systems
Yu Xiuwu, Li Ying, Liu Yong, Yu Hao
Summary: The IMCL-GABC algorithm optimizes deep mine localization using gray prediction and artificial bee colony algorithm, reducing localization error and time, and improving localization accuracy and efficiency.
Article
Computer Science, Information Systems
Seyed Salar Sefati, Mehrdad Abdi, Ali Ghaffari
Summary: Due to resource constraints, energy consumption and network lifetime are major challenges in wireless sensor networks (WSNs). This paper proposes a new routing scheme with load-balancing capability using the Markov Model (MM) and the Artificial Bee Colony (ABC) algorithm. The simulation results demonstrate that the proposed method outperforms other methods in terms of energy efficiency, number of alive nodes, and the number of delivered packets to BS and CH.
PEER-TO-PEER NETWORKING AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Deniz Ustun, Abdurrahim Toktas, Ugur Erkan, Ali Akdagli
Summary: This study introduces a modified algorithm, mABC, by incorporating mutation and crossover stages from differential evolution into the artificial bee colony algorithm. By comparing the performance with other algorithms through various statistical evaluations and convergence plots, it is found that mABC outperforms other variants in terms of optimization precision and convergence.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Chunlin He, Yong Zhang, Dunwei Gong, Xianfang Song, Xiaoyan Sun
Summary: In this article, an unsupervised multitask artificial bee colony (ABC) BS algorithm based on variable-size clustering (MBBS-VC) is proposed to simultaneously obtain multiple optimal band subsets with different sizes. Several new strategies are designed to improve the algorithm's performance, and experimental results verify the superiority of the proposed BS algorithm.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2022)
Article
Computer Science, Artificial Intelligence
Tingyu Ye, Wenjun Wang, Hui Wang, Zhihua Cui, Yun Wang, Jia Zhao, Min Hu
Summary: This article introduces a new artificial bee colony algorithm (RNSABC) based on random neighborhood structure to enhance the performance of the original ABC algorithm. The authors construct a random neighborhood structure and design an improved search strategy for optimization. Additionally, a depth-first search method is used to enhance the role of the onlooker bee phase. Experimental results demonstrate that RNSABC achieves competitive performance compared to nine other recent ABC variants.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Theory & Methods
Chao Lu, Xunbo Li, Wenjie Yu, Zhi Zeng, Mingming Yan, Xiang Li
Summary: This paper proposes a wireless sensor network coverage optimization method based on an improved artificial bee colony (ABC) algorithm, which combines the strong global search ability of ABC with the rapid convergence ability of TLBO, maintaining diversity and eliminating parameter limits.
Article
Computer Science, Information Systems
Sourabh Sharma, Harish Sharma, Janki Ballabh Sharma
Summary: This work proposes a perceptually tuned blind digital watermarking method for color images in a hybrid lifting wavelet transform and discrete cosine transform domain. The watermarked color image is encrypted using Arnold transform and embedded into the host image through quantization, DCT decomposition, and perceptually tuned dynamic embedding. The robustness and imperceptibility of the watermark are optimized using Artificial bee colony (ABC) optimization algorithm, showing effectiveness against image processing and manipulation attacks when compared with other methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Automation & Control Systems
Hanxiao Li, Kaizhou Gao, Pei-Yong Duan, Jun-Qing Li, Le Zhang
Summary: This work proposes an improved artificial bee colony algorithm with Q-learning, named QABC, for solving the permutation flow-shop scheduling problem. Experimental results demonstrate the superiority of QABC over other algorithms in solving the concerned problems.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Environmental Sciences
Jin Yan, Yuanyuan Chen, Jiazhu Zheng, Lin Guo, Siqi Zheng, Rongchun Zhang
Summary: This research utilizes advanced remote sensing technology and multi-source time series images to accurately extract and identify urban forests. The results show that the improved feature selection method has a high classification accuracy and is important for the management and monitoring of urban forests.
Article
Computer Science, Artificial Intelligence
Hui Wang, Wenjun Wang, Xinyu Zhou, Jia Zhao, Yun Wang, Songyi Xiao, Minyang Xu
Summary: KFABC is a novel ABC algorithm based on knowledge fusion, which enhances optimization capability by selecting and utilizing three kinds of knowledge, as well as adapting the learning mechanism based on search status. Results show that KFABC outperforms nine ABC and three differential evolution algorithms on thirty-two benchmark problems.
COMPLEX & INTELLIGENT SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Ali Ebrahimnejad, Mohammad Enayattabr, Homayun Motameni, Harish Garg
Summary: This study addresses the issue of shortest path problems with costs expressed in terms of mixed interval-valued fuzzy numbers by proposing a new algorithm. By approximating the summation of mixed interval-valued fuzzy numbers and introducing an extended distance function, the modified artificial bee colony algorithm is utilized to find the interval-valued membership function of the shortest path in such scenarios.
COMPLEX & INTELLIGENT SYSTEMS
(2021)
Article
Computer Science, Information Systems
Chouaib Ben Djaballah, Wahid Nouibat
Summary: The proposed multi-population ABC algorithm based on global and local optimum (MPGABC) outperforms other ABC algorithm variants in terms of competition and performance according to experimental results.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Jin Wang, Ying Liu, Shuying Rao, Xinyu Zhou, Jinbin Hu
Summary: This paper presents a novel self-adaptive multi-strategy artificial bee colony (SaMABC) algorithm for wireless sensor node coverage optimization. The algorithm designs an appropriate strategy pool and a fine-grained adaptive selection mechanism, and improves its optimization performance and ability to jump out of local optimums through the use of simulated annealing and dynamic search step.
Article
Computer Science, Information Systems
Yang Liu, Chaoqun Li, Yao Zhang, Mengying Xu, Jing Xiao, Jie Zhou
Summary: This study proposes a new clustering model named DCCM, based on duty cycle method, to reduce energy waste by enabling nodes to work alternately. An improved adaptive clone jellyfish search (DCC-IACJS) algorithm was designed to optimize the model. Simulation results demonstrate that DCC-IACJS outperforms other state-of-the-art counterparts in terms of network lifetime.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Computer Science, Hardware & Architecture
Yu Yang, Beifang Chen, Guoping Zhang, Yongming Li, Daoqiang Sun, Hongbo Liu
Summary: This paper proposes algorithms based on PCCWs, generating function, and structural decomposition to compute the subtree number index for graphs with more complicated cycle structures, such as bicyclic and tricyclic graphs. The approach provides a foundation and useful methods for investigating the structural properties of important nanomaterials.
Article
Automation & Control Systems
Dafeng Wang, Qian Ma, Naiyao Wang, Xuanzhe Fan, Mingyu Lu, Hongbo Liu
Summary: Predicting crowd flow accurately and timely is challenging due to uncertainty, but the proposed Active Offset Network (AONet) with ActiveGRU shows promise in predicting pedestrians' future positions. By utilizing convolution operation and bilinear interpolation, along with introducing a probabilistic sparse strategy, AONet outperforms state-of-art baselines in both accuracy and computational efficiency according to experiments on popular benchmarks.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Information Systems
Kai Liu, Hongbo Liu, Tomas E. Ward, Hua Wang, Yu Yang, Bo Zhang, Xindong Wu
Summary: The article introduces a novel method for detecting self-organized coalitions in brain functional networks, using the k-CLique Merging Evolution (CLIME) algorithm to embed cliques of each order k into a probabilistic mixture model, achieving robustness to density variation and coalition mixture. Experimental results demonstrate the effectiveness of the approach in detecting coalitions in brain functional networks.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2021)
Article
Computer Science, Artificial Intelligence
Bo Zhang, Naiyao Wang, Zheng Zhao, Ajith Abraham, Hongbo Liu
Summary: This paper presents an attention-guided multi-scale fusion network (AMS-Net) for crowd counting in dense scenarios, comprising density and attention networks. The proposed approach effectively handles persons of varied resolutions through a multi-scale fusion strategy built upon dilated convolution, as demonstrated by experimental results on standard benchmark datasets.
Article
Computer Science, Information Systems
Bo Zhang, Rui Zhang, Niccolo Bisagno, Nicola Conci, Francesco G. B. De Natale, Hongbo Liu
Summary: The proposed framework utilizes encoder-decoder and LSTM modules, along with social and physical attention mechanisms, to accurately predict individual and group behaviors in complex scenarios. Experiments demonstrate its effectiveness on standard crowd benchmarks.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2021)
Article
Automation & Control Systems
Bo Zhang, Chengzhi Yuan, Tao Wang, Hongbo Liu
Summary: This paper introduces a hybrid spatio-temporal embedding network (STENet) for human trajectory forecasting, incorporating 1D-CNN for position feature embedding and a two-stage graph attention mechanism for mutual interaction description. Group influences are taken into consideration with training done using the Wasserstein distance, showing effectiveness in experiments on ETH and UCY datasets.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Dafeng Wang, Hongbo Liu, Naiyao Wang, Yiyang Wang, Hua Wang, Sean McLoone
Summary: In this paper, a novel Sequence Entropy Energy-based Model (SEEM) is proposed to address the limitations of current trajectory prediction models. SEEM achieves diversity in candidate trajectory generation by optimizing sequence entropy, and improves accuracy and stability through probability distribution clipping mechanism and energy network. Experimental results demonstrate that SEEM outperforms the state-of-the-art approaches in terms of diversity, accuracy, and stability of pedestrian trajectory prediction.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Automation & Control Systems
Liping Yang, Yu Yang, Gervas Batister Mgaya, Bo Zhang, Long Chen, Hongbo Liu
Summary: Mining the relationship structures among investors is crucial for economic development and financial risk prevention in the era of big data. This article introduces fast networking approaches and novel algorithms to explore underlying structures in investment big data, showing higher clustering accuracy and efficiency compared to existing methods. Our method is particularly effective for mining investment structures.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Computer Science, Information Systems
Xin Jiang, Zhengxin Yu, Chao Hai, Hongbo Liu, Xindong Wu, Tomas Ward
Summary: This article proposes a transfer learning model called DNformer for predicting temporal link sequences in dynamic networks. By sequencing the structural dynamic evolution into consecutive links, capturing serial correlation using self-attention, and utilizing structural encoding to perceive the importance and correlation of links, the DNformer model outperforms other state-of-the-art TLP methods in various dynamic network problems.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2022)
Article
Automation & Control Systems
Bo Zhang, Tao Wang, Changdong Zhou, Nicola Conci, Hongbo Liu
Summary: This article introduces a flow-based multi-modal trajectory prediction framework that learns multi-modal distributions of trajectory data through an invertible network and generates plausible future paths. The framework explores advantages compared to GAN and VAE prediction frameworks.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Liping Yang, Xin Jiang, Yiming Ji, Hua Wang, Ajith Abraham, Hongbo Liu
Summary: This paper presents a novel gated graph convolutional network (GCN) based on spatio-temporal semi-variogram (STEM-GCN) for link prediction in dynamic networks. The approach learns spatial and temporal features and incorporates a correlation smoothing strategy to improve prediction accuracy. Experimental results demonstrate the effectiveness of the proposed approach.
Article
Computer Science, Artificial Intelligence
Zhenhao Shuai, Hongbo Liu, Zhaolin Wan, Wei-Jie Yu, Jun Zhang
Summary: This study proposes a self-adaptive neuroevolution (SANE) approach to automatically construct various lightweight DNN architectures for different tasks. SANE is able to self-adaptively adjust evolution exploration and exploitation to improve search efficiency. The results illustrate that the obtained DNN architectures could have smaller scale with similar performance compared to existing DNN architectures.
APPLIED SOFT COMPUTING
(2023)
Article
Engineering, Marine
Shaoyang Qiu, Hongxiang Ren, Naiyao Wang, Hongbo Liu
Summary: Studying the motion mechanism of a lifeboat's launch is crucial for ensuring marine evacuation safety. Motion modeling and simulation provide an effective approach to analyze the motion characteristics of lifeboat launches under different environmental conditions. However, current models for the complete launch process of a freefall lifeboat (FFLB) lack the latitudinal motion of both FFLB and its parent body, resulting in models with reduced accuracy and limited applicability. To address this issue, we propose a 3D motion model for FFLB launch from a moving ship using Kane's method. The model divides the launch process into sliding and water entry phases and considers the impact of ship motion and hydrodynamic forces on the lifeboat.
Article
Computer Science, Artificial Intelligence
Kai Liu, Hongbo Liu, Tao Wang, Guoqiang Hu, Tomas E. E. Ward, C. L. Philip Chen
Summary: A graph neural network (GNN) is a powerful architecture for semi-supervised learning. This article proposes a novel framework called graph coneighbor neural network (GCoNN) for node classification. GCoNN consists of two modules: GCoNN(G) and GCoNN(G)?. GCoNN(G) establishes the fundamental prototype for attribute learning while GCoNNi learns neighbor dependence through pseudolabels generated by GCoNN(G).
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Information Systems
Munan Li, Hongbo Liu, Xiangdong Jiang, Zheng Zhao, Tianhao Zhang
Summary: This paper proposes a novel unsupervised semantic learning model called SENSE for cross-platform vulnerability search. SENSE outperforms state-of-the-art methods in terms of binary search accuracy by pre-training semantic and graph learners and employing unsupervised batch-wise sampling with maximum mutual information loss.
COMPUTERS & SECURITY
(2023)