Article
Computer Science, Artificial Intelligence
Jin Huang, Tinghua Zhang, Weihao Yu, Jia Zhu, Ercong Cai
Summary: Community detection is a challenging and important problem in complex network analysis, and existing methods often overlook the overall characteristics and microscopic structure properties of the community. This paper proposes a novel model MDNMF, which can preserve both the topology information and intuitive structural properties of the community simultaneously, outperforming state-of-the-art approaches on well-known datasets.
INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Theory & Methods
Chaobo He, Hai Liu, Yong Tang, Shuangyin Liu, Xiang Fei, Qiwei Cheng, Hanchao Li
Summary: Community detection in signed networks is a challenging research problem, and overlapping community detection is a less explored direction. This paper proposes a similarity preserving overlapping community detection method (SPOCD) that fuses node similarity and geometric structure information to better preserve nodes with high similarity in the same community.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Computer Science, Information Systems
Sixing Su, Jiewen Guan, Bilian Chen, Xin Huang
Summary: This article proposes a community detection method based on node centrality under the framework of NMF. It designs a new similarity measure considering higher-order neighbors to form a more informative graph regularization mechanism and introduces node centrality and Gini impurity to measure the importance and sparseness of community memberships. Extensive experimental results demonstrate that the proposed method outperforms thirteen state-of-the-art methods on various real-world networks.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2023)
Article
Computer Science, Artificial Intelligence
Wentao Jia, Xiaoke Ma
Summary: This article proposes a novel algorithm called McNMF to detect the conserved communities in multi-layer networks. The algorithm simultaneously learns the shared and layer-specific features of vertices using joint non-negative matrix factorization, improving the quality of community structure. Extensive experiments show that the proposed algorithm outperforms state-of-the-art methods in terms of accuracy.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Xiaoxia Zhang, Xianjun Zhou, Lu Chen, Yanjun Liu
Summary: Explicable recommendation systems are important for improving the persuasiveness of the system and enhancing user trust. However, the presence of latent features makes it challenging to interpret recommendation results. To address this, a novel method called PE-NMF is proposed, which replaces latent variables with explicit data to help users understand the features of recommended items and make better decisions. Experimental results demonstrate that PE-NMF performs well in rating prediction and top-N recommendation, outperforming FE-NMF and maintaining comparable recommendation ability to NMF.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Engineering, Electrical & Electronic
Ying Zhang, Xiangli Li, Mengxue Jia
Summary: This paper introduces AGDNMF, an adaptive graph-based discriminative NMF method that utilizes label information to improve data representation and obtain the neighbor graph through adaptive iterations, which has been proven effective in various image datasets compared to state-of-the-art methods.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2021)
Article
Geochemistry & Geophysics
Ziyang Guo, Anyou Min, Bing Yang, Junhong Chen, Hong Li, Junbin Gao
Summary: This article proposes a new sparse oblique-manifold (OB) NMF method inspired by matrix manifold theory, treating the abundance matrix as located on the oblique manifold to eliminate constraints and incorporate intrinsic Riemannian geometry. By using the Riemannian conjugated gradient (RCG) algorithm and multiplicative iterative rule, the proposed method not only improves solution accuracy but also achieves a faster convergence rate. Experimental results demonstrate the effectiveness and efficiency of the proposed method compared to state-of-the-art NMF methods in HU.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Interdisciplinary Applications
Zhenhai Chang, Zhong-Yuan Zhang, Huimin Cheng, Chao Yan, Xianjun Yin
Summary: The paper discusses a community structure detection method based on nonnegative matrix factorization, and proposes a new algorithm that reformulates modularity maximization under the Frobenius norm framework. Experimental results show that the new method outperforms existing methods in clustering quality.
INTERNATIONAL JOURNAL OF MODERN PHYSICS C
(2021)
Article
Computer Science, Information Systems
Siyuan Peng, Zhijing Yang, Bingo Wing-Kuen Ling, Badong Chen, Zhiping Lin
Summary: A new semi-supervised NMF method called dual semi-supervised convex nonnegative matrix factorization (DCNMF) is proposed in this paper. DCNMF incorporates the pointwise and pairwise constraints of labeled samples into convex NMF, resulting in a better low-dimensional data representation. It can process mixed-sign data due to the nonnegative constraint only on the coefficient matrix.
INFORMATION SCIENCES
(2022)
Article
Engineering, Electrical & Electronic
Mikael Sorensen, Nicholas D. Sidiropoulos, Ananthram Swami
Summary: A new SBMF model is proposed for community detection, which can handle overlapping communities effectively, and its effectiveness is demonstrated through experiments.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2022)
Article
Computer Science, Information Systems
Xiaoxia Zhang, Degang Chen, Hong Yu, Guoyin Wang, Houjun Tang, Kesheng Wu
Summary: Nonnegative Matrix Factorization (NMF) produces interpretable solutions for applications like collaborative filtering. Regularization is needed to address issues like overfitting and interpretability. Existing regularizers are constructed from factorization results, but this study proposes a more holistic graph regularizer based on a linear projection of the rating matrix, named LPGNMF. Experimental results show the superiority of LPGNMF on different datasets.
INFORMATION SCIENCES
(2022)
Article
Mathematics
Yunxia Xu, Linzhang Lu, Qilong Liu, Zhen Chen
Summary: In this paper, a hypergraph-regularized Lp smooth nonnegative matrix factorization (HGSNMF) is proposed by incorporating hypergraph regularization and Lp smoothing constraint into standard NMF model. The method captures the intrinsic geometry structure of high dimension space data comprehensively and provides a smooth and accurate solution to the optimization problem. Experimental results show that the proposed method outperforms the state-of-the-art methods in most cases.
Article
Computer Science, Artificial Intelligence
Mingyang Liu, Zuyuan Yang, Lingjiang Li, Zhenni Li, Shengli Xie
Summary: Multi-view clustering is an attractive approach that combines information from multiple views. The collective matrix factorization (CMF) has been proved to be effective in extracting shared information for multi-view data. In this study, we propose a novel unified multi-view clustering framework, ACMF-GDR, which employs auto-weighted CMF with graph dual regularization. Experimental results demonstrate the superior performance of the proposed method in clustering.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Ping Deng, Tianrui Li, Hongjun Wang, Shi-Jinn Horng, Zeng Yu, Xiaomin Wang
Summary: The objective of co-clustering is to identify similarity blocks between the sample set and feature set simultaneously. The nonnegative matrix tri-factorization algorithm is an important tool for co-clustering. To address the impact of noise, a tri-regularized NMTF model was proposed to optimize the performance and generalization ability of the model effectively.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Junjun Pan, Nicolas Gillis
Summary: Nonnegative matrix factorization (NMF) is a linear dimensionality technique for nonnegative data, which can be efficiently computed under the separability assumption. The algorithm operates by finding data points that contain basis vectors for decomposition.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Physics, Multidisciplinary
Zhenhai Chang, Xianjun Yin, Caiyan Jia, Xiaoyang Wang
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2018)
Article
Physics, Multidisciplinary
Chao Yan, Zhenhai Chang
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2019)
Article
Computer Science, Information Systems
Zhenhai Chang, Caiyan Jia, Xianjun Yin, Yimei Zheng
INFORMATION SCIENCES
(2019)
Article
Physics, Multidisciplinary
Xiaoyu Shi, Jian Zhang, Xia Jiang, Juan Chen, Wei Hao, Bo Wang
Summary: This study presents a novel framework using offline reinforcement learning to improve energy consumption in road transportation. By leveraging real-world human driving trajectories, the proposed method achieves significant improvements in energy consumption. The offline learning approach demonstrates generalizability across different scenarios.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Junhyuk Woo, Soon Ho Kim, Hyeongmo Kim, Kyungreem Han
Summary: Reservoir computing (RC) is a new machine-learning framework that uses an abstract neural network model to process information from complex dynamical systems. This study investigates the neuronal and network dynamics of liquid state machines (LSMs) using numerical simulations and classification tasks. The findings suggest that the computational performance of LSMs is closely related to the dynamic range, with a larger dynamic range resulting in higher performance.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Yuwei Yang, Zhuoxuan Li, Jun Chen, Zhiyuan Liu, Jinde Cao
Summary: This paper proposes an extreme learning machine (ELM) algorithm based on residual correction and Tent chaos sequence (TRELM-DROP) for accurate prediction of traffic flow. The algorithm reduces the impact of randomness in traffic flow through the Tent chaos strategy and residual correction method, and avoids weight optimization using the iterative method. A DROP strategy is introduced to improve the algorithm's ability to predict traffic flow under varying conditions.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Chengwei Dong, Min Yang, Lian Jia, Zirun Li
Summary: This work presents a novel three-dimensional system with multiple types of coexisting attractors, and investigates its dynamics using various methods. The mechanism of chaos emergence is explored, and the periodic orbits in the system are studied using the variational method. A symbolic coding method is successfully established to classify the short cycles. The flexibility and validity of the system are demonstrated through analogous circuit implementation. Various chaos-based applications are also presented to show the system's feasibility.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Viorel Badescu
Summary: This article discusses the maximum work extraction from confined particles energy, considering both reversible and irreversible processes. The results vary for different types of particles and conditions. The concept of exergy cannot be defined for particles that undergo spontaneous creation and annihilation. It is also noted that the Carnot efficiency is not applicable to the conversion of confined thermal radiation into work.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
P. M. Centres, D. J. Perez-Morelo, R. Guzman, L. Reinaudi, M. C. Gimenez
Summary: In this study, a phenomenological investigation of epidemic spread was conducted using a model of agent diffusion over a square region based on the SIR model. Two possible contagion mechanisms were considered, and it was observed that the number of secondary infections produced by an individual during its infectious period depended on various factors.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Zuan Jin, Minghui Ma, Shidong Liang, Hongguang Yao
Summary: This study proposes a differential variable speed limit (DVSL) control strategy considering lane assignment, which sets dynamic speed limits for each lane to attract vehicle lane-changing behaviors before the bottleneck and reduce the impact of traffic capacity drop. Experimental results show that the proposed DVSL control strategy can alleviate traffic congestion and improve efficiency.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Matthew Dicks, Andrew Paskaramoorthy, Tim Gebbie
Summary: In this study, we investigate the learning dynamics of a single reinforcement learning optimal execution trading agent when it interacts with an event-driven agent-based financial market model. The results show that the agents with smaller state spaces converge faster and are able to intuitively learn to trade using spread and volume states. The introduction of the learning agent has a robust impact on the moments of the model, except for the Hurst exponent, which decreases, and it can increase the micro-price volatility as trading volumes increase.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Zhouzhou Yao, Xianyu Wu, Yang Yang, Ning Li
Summary: This paper developed a cooperative lane-changing decision system based on digital technology and indirect reciprocity. By introducing image scoring and a Q-learning based reinforcement learning algorithm, drivers can continuously evaluate gains and adjust their strategies. The study shows that this decision system can improve driver cooperation and traffic efficiency, achieving over 50% cooperation probability under any connected vehicles penetration and traffic density, and reaching 100% cooperation probability under high penetration and medium to high traffic density.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Josephine Nanyondo, Henry Kasumba
Summary: This paper presents a multi-class Aw-Rascle (AR) model with area occupancy expressed in terms of vehicle class proportions. The qualitative properties of the proposed equilibrium velocity and the stability conditions of the model are established. The numerical results show the effect of proportional densities on the flow of vehicle classes, indicating the realism of the proposed model.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Oliver Smirnov
Summary: This study proposes a new method for simultaneously estimating the parameters of the 2D Ising model. The method solves a constrained optimization problem, where the objective function is a pseudo-log-likelihood and the constraint is the Hamiltonian of the external field. Monte Carlo simulations were conducted using models of different shapes and sizes to evaluate the performance of the method with and without the Hamiltonian constraint. The results demonstrate that the proposed estimation method yields lower variance across all model shapes and sizes compared to a simple pseudo-maximum likelihood.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Przemyslaw Chelminiak
Summary: The study investigates the first-passage properties of a non-linear diffusion equation with diffusivity dependent on the concentration/probability density through a power-law relationship. The survival probability and first-passage time distribution are determined based on the power-law exponent, and both exact and approximate expressions are derived, along with their asymptotic representations. The results pertain to diffusing particles that are either freely or harmonically trapped. The mean first-passage time is finite for the harmonically trapped particle, while it is divergent for the freely diffusing particle.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Hidemaro Suwa
Summary: The choice of transition kernel is crucial for the performance of the Markov chain Monte Carlo method. A one-parameter rejection control transition kernel is proposed, and it is shown that the rejection process plays a significant role in determining the sampling efficiency.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)
Article
Physics, Multidisciplinary
Xudong Wang, Yao Chen
Summary: This article investigates the joint influence of expanding medium and constant force on particle diffusion. By starting from the Langevin picture and introducing the effect of external force in two different ways, two models with different force terms are obtained. Detailed analysis and derivation yield the Fokker-Planck equations and moments for the two models. The sustained force behaves as a decoupled force, while the intermittent force changes the diffusion behavior with specific effects depending on the expanding rate of the medium.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2024)