Search

Graph clustering is a critical task in network analysis, aimed at grouping nodes based on their structural or attributed similarities. In particular, attributed graph clustering, which considers both structural links and node attributes, is essential for complex networks where additional node information is available. Nonnegative Matrix Factorization (NMF) has shown promise in graph clustering; however, it faces limitations when applied to attributed graph clustering, such as an inability to detect outliers, distortion of geometric data point structures, and disregard for attributed information. Moreover, many existing attributed graph clustering methods overlook distant node relationships due to network sparsity, which hinders further performance improvements. To address these challenges, this paper introduces Weighted Symmetric NMF with graph-Boosting for attributed Graph Clustering (WSBGC), an innovative extension of NMF. WSBGC leverages attribute similarity among nodes to create a weighted version of NMF, enabling outlier detection while preserving the geometric structure of data points. Additionally, WSBGC employs graph-boosting, leveraging attribute information to account for distant node relationships and improve clustering accuracy. A graph attention auto-encoder is then used to construct the final clustering model. The effectiveness of WSBGC is validated through extensive experiments on real-world datasets. Notably, our algorithm improves accuracy by 2.5% compared to the best available method, demonstrating its superior clustering performance in attributed graphs. • Using graph-boosting and nonnegative matrix factorization to improve graph clustering. • A weighted symmetric model of NMF for attributed graph clustering is introduced. • Graph-boosting technique is applied to increase cluster quality. • Nonnegative matrix factorization is applied to capture the hidden structures in the data. • Our algorithm incorporates distant node relationships into the clustering process. [ABSTRACT FROM AUTHOR]

• Description of consensus control of physically coupled MASs,. • Sufficient condition for consensusablility of physically interconnected MASs. • Constrained modelling of cable suspended load transportation system. This study focuses on achieving optimal consensus control based on the state-dependent Riccati equation (SDRE) in multi-agent systems (MASs) characterized by the interconnectivity of multiple sub-systems with physical interconnections. The objective is to establish a suitable communication topology among the MAS controllers that considers both local and global objectives while ensuring stability through distributed control. To address this, an innovative extension of leader-follower and leaderless consensus control methodologies is proposed, accompanied by stability analysis employing the Lyapunov method. Additionally, the paper explores consensus control for dynamically interconnected MASs handling suspended load transportation, such as multi-lift systems. By utilizing the principles of virtual work by D'Alembert and Gauss's principle of least constraint, a constrained multibody system model is developed, incorporating the Udwadia-Kalaba motion equation approach. Simulation results and theoretical analysis validate the efficiency and performance of the proposed consensus controller structure. [ABSTRACT FROM AUTHOR]

• SegMENT-Plus automatically recognizes laryngeal tumor margins in WL and NBI images. • SegMENT-Plus includes EfficientNetB5, m-ASPP, CBAM for precise cancer delineation. • Validated on three hospitals, with outcome 81.4 to 84.9% DSC and 81.8 to 85.7% IOU. • The m-ASPP surpassed stranded ASPP, WASPP and Dense ASPP. • Fusion of CBAM and SE: This combination improves laryngeal tumor sensitivity. White Light (WL) and Narrow Band Imaging (NBI) endoscopy are widely used to assess the superficial spreading of laryngeal squamous cell carcinoma (LSCC). However, the analysis of images requires a high level of attention and extensive clinical expertise, leading to inter-clinician variability on the assessment of tumor margins. Computer-aided segmentation can automate the identification of LSCC margins, supporting clinicians in this challenging task. In this paper, we present SegMENT-Plus , a Deep Learning segmentation convolutional network specifically developed and optimized for accurate delineation of LSCC. SegMENT-Plus uses EfficienstNetB5 as encoder with a new modified Atrous Spatial Pyramid Pooling (m-ASPP) block that integrates Channel Block Attention Module (CBAM) and Squeeze Excitation (SE). In this new architecture, CBAM extracts local and global LSCC features from the encoder, while the SE block refines cancer segmentation on each dilated convolution output. SegMENT-Plus was trained and evaluated on a multi-center dataset including clinical data from three different hospitals. A total of 4289 annotated laryngeal images from 766 patients were included in this study. The experiments showed that SegMENT-Plus achieved a Dice Similarity Coefficient (DSC) between 81.4% and 84.9% and an Intersection over Union (IOU) between 81.8% and 85.7% on the data from the different hospitals, attesting its high performance and generalization capability. The proposed segmentation architecture also demonstrated statistically significant improvement in DSC and IoU compared to other state of the art architectures, showing that this work is a concrete foundation towards a clinical system for the automatic delineation of LSCC margins in endoscopic images. [ABSTRACT FROM AUTHOR]

Clustering is a fundamental task in data analysis, aiming to partition a set of samples into subsets called clusters based on their similarity. Agglomerative Hierarchical Clustering (AHC) is a common hierarchical clustering method in which clusters are formed bottom-up. The AHC-based ensemble clustering technique has emerged as an important extension of classical clustering problems and is one of the most recent advances in unsupervised learning. The aim of this technique is to combine the output results of several individual AHC-based methods to achieve a consensus with higher accuracy. In this study, we address the problem of semi-supervised ensemble clustering, which combines ensemble clustering methods and semi-supervised clustering techniques to improve clustering accuracy. We propose a novel framework that integrates AHC methods to create primary partitions, leveraging the diversity offered by AHC to enhance clustering quality. Our approach includes a comprehensive selection process to reduce computational complexity and a new similarity measure developed using a flexible weighting mechanism. The purpose of this mechanism is to describe the consistency between semi-supervised clustering methods used to generate base partitions. In general, the three main steps of the proposed method include creating primary clusters with different AHC methods based on constraint information, developing an innovative similarity measure to calculate the similarity between objects, and finally re-clustering the primary clusters to create final clusters. We conducted experiments on real-world datasets to evaluate the effectiveness of the proposed method. The findings show significant improvements in clustering accuracy compared to equivalent methods. • Presenting a robust semi-supervised method based on ensemble learning. • Configuration of an AHC-based clustering framework by joining ECM and SSC. • Introducing a new similarity measure based on a flexible weighting mechanism. • Development of a merit-based selection strategy to reduce complexity in consensus. • The proposed method performs well in sparse and large-scale data. [ABSTRACT FROM AUTHOR]