Semi-supervised Spectral Clustering Research Articles

Semi-supervised spectral clustering using shared nearest neighbour for data with different shape and density

<p>In the absence of supervisory information in spectral clustering algorithms, it is difficult to construct suitable similarity graphs for data with complex shapes and varying densities. To address this issue, this paper proposes a Semi-supervised Spectral Clustering algorithm based on shared nearest neighbor. The proposed algorithm combines the idea of semi-supervised clustering, adding Shared Nearest Neighbor information to the calculation of the distance matrix, and using pairwise constraint information to find the relationship between two data points, while providing a portion of supervised information. Comparative experiments were conducted on artificial data sets and University of California Irvine machine learning repository datasets. The experimental results show that the proposed algorithm achieves better clustering results compared to traditional K-means and spectral clustering algorithms.</p>

Consumer-feeder relationship identification method for low-voltage distribution station area considering feature fusion of voltage variation and spatial distribution

Consumer-feeder relationship is an important basis for the fault location, outage management, load adjustment and power quality management in low-voltage distribution station area (LVDSA). The current identification method of the consumer-feeder relationship based on electrical characteristics analysis is single and lack of reliability. Besides, the consumers of different feeders which locate near the bus at the low voltage side of distribution transformer have high voltage variation similarity, which makes it difficult to distinguish reliably. The spatial distribution of consumers can directly reflects the feeder trend. Therefore, a consumer-feeder relationship identification method for LVDSA considering feature fusion of voltage variation and spatial distribution is proposed in this paper. Firstly, the electrical characteristics based on voltage variation and spatial distribution characteristics based on geographical location are analyzed, and the correlation between the characteristics above and consumer-feeder relationship is explained. Secondly, a semi-supervised spectral clustering (SSC) algorithm based on the feature of voltage variation and spatial distribution is proposed to identify the consumer-feeder relationship. Finally, the proposed method is tested with the real data of LVDSAs in China. The results indicate that the feature fusion of voltage variation and spatial distribution can effectively improve the identification accuracy of consumer-feeder relationship.

Graph clustering and variational image segmentation for automated firearm detection in X‐ray images

Millions of parcels and pieces of luggage are scanned daily for threat detection at border control or high-security buildings. Currently, the process is manually operated by security agents and it is slow and time-consuming. The automation of this process will lift the burden from the security agents and will allow larger volumes of items to be scanned. The authors consider the problem of automatic threat detection, in particular firearms, in X-rays. To achieve this goal, they propose a hybrid algorithm that combines two well-established image segmentation algorithms into a two-step clustering method. The first step is a semi-supervised spectral clustering algorithm at the image level, which classifies whole images into benign or containing a threat. The images classified as threatening from the first step proceed to the second stage, where a variational image segmentation algorithm performs clustering at the pixel level to locate the threat if it exists. The hybrid algorithm is designed to scale-up the processing of hundreds of images, in comparison to the academic literature where only a handful images are used for demonstration. Numerical experiments establish that the combination of two different algorithms produces better results than using individual algorithms.

SolidBin: improving metagenome binning with semi-supervised normalized cut.

Metagenomic contig binning is an important computational problem in metagenomic research, which aims to cluster contigs from the same genome into the same group. Unlike classical clustering problem, contig binning can utilize known relationships among some of the contigs or the taxonomic identity of some contigs. However, the current state-of-the-art contig binning methods do not make full use of the additional biological information except the coverage and sequence composition of the contigs. We developed a novel contig binning method, Semi-supervised Spectral Normalized Cut for Binning (SolidBin), based on semi-supervised spectral clustering. Using sequence feature similarity and/or additional biological information, such as the reliable taxonomy assignments of some contigs, SolidBin constructs two types of prior information: must-link and cannot-link constraints. Must-link constraints mean that the pair of contigs should be clustered into the same group, while cannot-link constraints mean that the pair of contigs should be clustered in different groups. These constraints are then integrated into a classical spectral clustering approach, normalized cut, for improved contig binning. The performance of SolidBin is compared with five state-of-the-art genome binners, CONCOCT, COCACOLA, MaxBin, MetaBAT and BMC3C on five next-generation sequencing benchmark datasets including simulated multi- and single-sample datasets and real multi-sample datasets. The experimental results show that, SolidBin has achieved the best performance in terms of F-score, Adjusted Rand Index and Normalized Mutual Information, especially while using the real datasets and the single-sample dataset. https://github.com/sufforest/SolidBin. Supplementary data are available at Bioinformatics online.

Controlled Islanding for a Hybrid AC/DC Grid With VSC-HVDC Using Semi-Supervised Spectral Clustering

As the last resort of emergency control, controlled islanding is an effective means of preventing fault-propagation and a system-wide blackout. However, conventional AC transmission lines are unavailable to be employed for power exchange between these islands. To make full use of the DC power modulation capability of VSC-HVDC links, a new controlled islanding model is put forward for an AC/VSC-HVDC hybrid grid to minimize the composite power-flow disruption, in which the DC-terminals belonging to a VSC-HVDC link are placed in different islands. To solve this model, a semi-supervised spectral clustering-based approach is proposed by transforming the problem into a weighted undirected graph segmentation problem. The novelty of our work is to find an optimal islanding solution in real time such that the power exchanges between islands are implemented via a VSC-HVDC link to reduce the generation-load imbalance. The simulation results on the IEEE 39-bus system and a real-world system verify the effectiveness and superiority of the proposed approach.

Unsupervised TOF Image Segmentation through Spectral Clustering and Region Merging

Time of Flight (TOF) cameras generate two simultaneous images, one of intensity and one of range. This allows to tackle segmentation problems in which the separate use of intensity or range information is not enough to extract objects of interest from the 3D scene. In turn, range information allows to obtain a normal vector estimation of each point of the captured surfaces. This article presents a semi-supervised spectral clustering method which combines intensity and range information as well as normal vector orientations to improve segmentation results. The main contribution of this article consists in the use of a statistical region merging as a final step of the segmentation method. The region merging process combines adjacent regions which satisfy a similarity criterion. The performance of the proposed method was evaluated over real images. The use of this final step presents preliminary improvements in the metrics evaluated.

Individualized Functional Parcellation of the Human Amygdala Using a Semi-supervised Clustering Method: A 7T Resting State fMRI Study.

The amygdala plays an important role in emotional functions and its dysfunction is considered to be associated with multiple psychiatric disorders in humans. Cytoarchitectonic mapping has demonstrated that the human amygdala complex comprises several subregions. However, it's difficult to delineate boundaries of these subregions in vivo even if using state of the art high resolution structural MRI. Previous attempts to parcellate this small structure using unsupervised clustering methods based on resting state fMRI data suffered from the low spatial resolution of typical fMRI data, and it remains challenging for the unsupervised methods to define subregions of the amygdala in vivo. In this study, we developed a novel brain parcellation method to segment the human amygdala into spatially contiguous subregions based on 7T high resolution fMRI data. The parcellation was implemented using a semi-supervised spectral clustering (SSC) algorithm at an individual subject level. Under guidance of prior information derived from the Julich cytoarchitectonic atlas, our method clustered voxels of the amygdala into subregions according to similarity measures of their functional signals. As a result, three distinct amygdala subregions can be obtained in each hemisphere for every individual subject. Compared with the cytoarchitectonic atlas, our method achieved better performance in terms of subregional functional homogeneity. Validation experiments have also demonstrated that the amygdala subregions obtained by our method have distinctive, lateralized functional connectivity (FC) patterns. Our study has demonstrated that the semi-supervised brain parcellation method is a powerful tool for exploring amygdala subregional functions.

Semi-Supervised Spectral Clustering With Structured Sparsity Regularization

Spectral clustering (SC) is one of the most widely used clustering methods. In this letter, we extend the traditional SC with a semi-supervised manner. Specifically, with the guidance of small amount of supervisory information, we build a matrix with anti-block-diagonal appearance, which is further utilized to regularize the product of the low-dimensional embedding and its transpose. Technically, we formulate the proposed model as a constrained optimization problem. Then, we relax it as a convex problem, which can be efficiently solved with the global convergence guaranteed via the inexact augmented Lagrangian multiplier method. Experimental results over four real-world datasets demonstrate that higher accuracy and normalized mutual information are achieved when compared with state-of-the-art methods.

Constrains optimal propagation-based modified semi-supervised spectral clustering for large-scale data

We focus on the problem of high computational complexity in the clustering process of traditional spectral clustering algorithm that cannot satisfy the requirement of current large-scale data clustering applications. In this article, we establish a constrained optimal propagation based semi-supervised large-scale data clustering model. In this model, micro similarity matrix is constructed by using prior dotted pair constraint information at first. On this basis, the Gabow algorithm is exploited to extract each strongly connected component from the micro similarity matrix that is represented by its connected graph. Then, a new constrained optimisation propagation algorithm for each strongly connected component is proposed to calculate the similarity of the whole dataset. Finally, we employ the singular value decomposition and the accelerated k-means algorithm to obtain the clustering results of large-scale data. Experiments on multiple standard testing datasets show that compared with other previous research results in this field, the proposed clustering model has higher clustering accuracy and lower computation complexity, and is more suitable for large-scale data clustering applications.

Constrains optimal propagation-based modified semi-supervised spectral clustering for large-scale data

Constrains optimal propagation-based modified semi-supervised spectral clustering for large-scale data

Semi-supervised spectral algorithms for community detection in complex networks based on equivalence of clustering methods

Community detection is fundamental for revealing the structure-functionality relationship in complex networks, which involves two issues—the quantitative function for community as well as algorithms to discover communities. Despite significant research on either of them, few attempt has been made to establish the connection between the two issues. To attack this problem, a generalized quantification function is proposed for community in weighted networks, which provides a framework that unifies several well-known measures. Then, we prove that the trace optimization of the proposed measure is equivalent with the objective functions of algorithms such as nonnegative matrix factorization, kernel K-means as well as spectral clustering. It serves as the theoretical foundation for designing algorithms for community detection. On the second issue, a semi-supervised spectral clustering algorithm is developed by exploring the equivalence relation via combining the nonnegative matrix factorization and spectral clustering. Different from the traditional semi-supervised algorithms, the partial supervision is integrated into the objective of the spectral algorithm. Finally, through extensive experiments on both artificial and real world networks, we demonstrate that the proposed method improves the accuracy of the traditional spectral algorithms in community detection.

Self-weighted spectral clustering with parameter-free constraint

The constrained spectral clustering (or known as the semi-supervised spectral clustering) focuses on enhancing the clustering capability by utilizing the side information. In this paper, a novel constrained spectral clustering method is proposed based on deriving a sparse parameter-free similarity. Different from other works, the proposed method transforms the given pairwise constraints into the intrinsic graph similarity and the penalty graph similarity respectively instead of incorporating them into one single similarity. Besides, the optimal weight can be automatically achieved to balance the graph optimization problems between the intrinsic graph and the penalty graph. Equipped with a general framework of efficiently unraveling the bi-objective optimization, the proposed method could obtain both ratio cut and normalized cut clusterings via updating the weighted Laplacian matrix until convergence. Moreover, the proposed method is equivalent to the spectral clustering, when no side information is provided. Consequently, the effectiveness and the superiority of the proposed method are further verified both analytically and empirically.

A semi-supervised locally linear embedding spectral clustering algorithm

Many practical problems can be attributed to the clustering problem. Spectral clustering algorithm can be clustered in any shape of space, and obtain the global optimal solution. Based on the classical Ng-Jordan-Weiss (NJW) algorithm, utilising the supervision information to guide the clustering process, the result of clustering is more accurate. Meanwhile, combined the manifold learning with semi-supervised spectral clustering algorithm, and the data dimension will reduce based on locally linear embedding (LLE). Based on the heuristic thinking, calculated distance matrix, a reasonable number of nearest neighbours could be funded, thus we achieve the purpose of dimension reduction. Moreover, clustering based on reduced dimension data, the same clustering results as the original data could be obtained. Experimental results have shown that this algorithm could achieve better clustering effect on artificial datasets and real datasets.

A semi-supervised locally linear embedding spectral clustering algorithm

Many practical problems can be attributed to the clustering problem. Spectral clustering algorithm can be clustered in any shape of space, and obtain the global optimal solution. Based on the classical Ng-Jordan-Weiss (NJW) algorithm, utilising the supervision information to guide the clustering process, the result of clustering is more accurate. Meanwhile, combined the manifold learning with semi-supervised spectral clustering algorithm, and the data dimension will reduce based on locally linear embedding (LLE). Based on the heuristic thinking, calculated distance matrix, a reasonable number of nearest neighbours could be funded, thus we achieve the purpose of dimension reduction. Moreover, clustering based on reduced dimension data, the same clustering results as the original data could be obtained. Experimental results have shown that this algorithm could achieve better clustering effect on artificial datasets and real datasets.

Active Clustering with Model-Based Uncertainty Reduction.

Semi-supervised clustering seeks to augment traditional clustering methods by incorporating side information provided via human expertise in order to increase the semantic meaningfulness of the resulting clusters. However, most current methods are passive in the sense that the side information is provided beforehand and selected randomly. This may require a large number of constraints, some of which could be redundant, unnecessary, or even detrimental to the clustering results. Thus in order to scale such semi-supervised algorithms to larger problems it is desirable to pursue an active clustering method-i.e., an algorithm that maximizes the effectiveness of the available human labor by only requesting human input where it will have the greatest impact. Here, we propose a novel online framework for active semi-supervised spectral clustering that selects pairwise constraints as clustering proceeds, based on the principle of uncertainty reduction. Using a first-order Taylor expansion, we decompose the expected uncertainty reduction problem into a gradient and a step-scale, computed via an application of matrix perturbation theory and cluster-assignment entropy, respectively. The resulting model is used to estimate the uncertainty reduction potential of each sample in the dataset. We then present the human user with pairwise queries with respect to only the best candidate sample. We evaluate our method using three different image datasets (faces, leaves and dogs), a set of common UCI machine learning datasets and a gene dataset. The results validate our decomposition formulation and show that our method is consistently superior to existing state-of-the-art techniques, as well as being robust to noise and to unknown numbers of clusters.

Using automated walking gait analysis for the identification of pedestrian attributes

Collecting microscopic pedestrian behavior and characteristics data is important for optimizing the design of pedestrian facilities for safety, efficiency, and comfortability. This paper provides a framework for the automated classification of pedestrian attributes such as age and gender based on information extracted from their walking gait behavior. The framework extends earlier work on the automated analysis of gait parameters to include analysis of the gait acceleration data which can enable the quantification of the variability, rhythmic pattern and stability of pedestrian’s gait. In this framework, computer vision techniques are used for the automatic detection and tracking of pedestrians in an open environment resulting in pedestrian trajectories and the speed and acceleration dynamic profiles. A collection of gait features are then derived from those dynamic profiles and used for the classification of pedestrian attributes. The gait features include conventional gait parameters such as gait length and frequency and dynamic parameters related to gait variations and stability measures. Two different techniques are used for the classification: a supervised k-Nearest Neighbors (k-NN) algorithm and a newly developed semi-supervised spectral clustering. The classification framework is demonstrated with two case studies from Vancouver, British Columbia and Oakland, California. The results show the superiority of features sets including gait variations and stability measures over features relying only on conventional gait parameters. For gender, correct classification rates (CCR) of 80% and 94% were achieved for the Vancouver and Oakland case studies, respectively. The classification accuracy for gender was higher in the Oakland case which only considered pedestrians walking alone. Pedestrian age classification resulted in a CCR of 90% for the Oakland case study.

Semi-supervised spectral clustering with application to detect population stratification

In genetic association studies, unaccounted population stratification can cause spurious associations in a discovery process of identifying disease-associated genetic markers. In such a situation, prior information is often available for some subjects' population identities. To leverage the additional information, we propose a semi-supervised clustering approach for detecting population stratification. This approach maintains the advantages of spectral clustering, while is integrated with the additional identity information, leading to sharper clustering performance. To demonstrate utility of our approach, we analyze a whole-genome sequencing dataset from the 1000 Genomes Project, consisting of the genotypes of 607 individuals sampled from three continental groups involving 10 subpopulations. This is compared against a semi-supervised spectral clustering method, in addition to a spectral clustering method, with the known subpopulation information by the Rand index and an adjusted Rand (ARand) index. The numerical results suggest that the proposed method outperforms its competitors in detecting population stratification.

Research of semi-supervised spectral clustering algorithm based on pairwise constraints

Clustering is often considered as an unsupervised data analysis method, but making full use of the prior information in the process of clustering will significantly improve the performance of the clustering algorithm. Spectral clustering algorithm can well use the prior pairwise constraint information to cluster and has become a new hot spot of machine learning research in recent years. In this paper, we propose an effective clustering algorithm, called a semi-supervised spectral clustering algorithm based on pairwise constraints, in which the similarity matrix of data points is adjusted and optimized by pairwise constraints. The experiments on real-world data sets demonstrate the effectiveness of this algorithm.

Research of semi-supervised spectral clustering based on constraints expansion

Semi-supervised learning has become one of the hotspots in the field of machine learning in recent years. It is successfully applied in clustering and improves the clustering performance. This paper proposes a new clustering algorithm, called semi-supervised spectral clustering based on constraints expansion (SSCCE). This algorithm expands the known constraints set, changes the similarity relation of the sample points through the density–sensitive path distance, and then combines with semi-supervised spectral clustering to cluster. The experimental results prove that SSCCE algorithm has good clustering effect.

Spectral clustering: A semi-supervised approach

Recently, graph-based spectral clustering algorithms have been developing rapidly, which are proposed as discrete combinatorial optimization problems and approximately solved by relaxing them into tractable eigenvalue decomposition problems. In this paper, we first review the current existing spectral clustering algorithms in a unified-framework way and give a straightforward explanation about spectral clustering. We also present a novel model for generalizing the unsupervised spectral clustering to semi-supervised spectral clustering. Under this model, prior information given by some instance-level constraints can be generalized to space-level constraints. We find that (undirected) graph built on the enlarged prior information is more meaningful, hence the boundaries of the clusters are more correct. Experimental results based on toy data, real-world data and image segmentation demonstrate the advantages of the proposed model.