Affinity Propagation Clustering Algorithm Research Articles

A Novel Self-Adaptive Affinity Propagation Clustering Algorithm Based on Density Peak Theory and Weighted Similarity

A Novel Self-Adaptive Affinity Propagation Clustering Algorithm Based on Density Peak Theory and Weighted Similarity

A feature weighted affinity propagation clustering algorithm based on rough entropy reduction

In the clustering task, each feature of data sample is not taken the same contribution, some features provides more related information to the final results, if they are treated equally with other features, not only the complexity of the algorithm is increased but also the accuracy of the final results will be affected. So as a key phase in clustering, feature weighting is becoming more and more concerned by scholars. This paper proposes a feature weighted affinity propagation clustering algorithm based on rough entropy reduction (FWRER-AP). Rough entropy is used to assign weights for every feature according to their different contribution. Then the optimisation samples are used in AP clustering algorithm, we can get the final clustering results through iterations. Compared with traditional AP clustering algorithm, experiment shows that the optimal algorithm not only reduces the complexity, but also improves the accuracy at the same time.

Multi-label learning method based on ML-RBF and laplacian ELM

Multi-label data widely exist in the real world, and the multi-label learning deals with the problem in which samples contain many labels. The main task of the multi-label learning is to train a model which adapts to the multi-label data such that the label of unknown label data can be predicted. Multi-label radial basis function neural network (ML-RBF) is an effective multi-label learning model, which combines K-means clustering and RBF neural network. The laplacian extreme learning machine (Lap-ELM) improved the traditional ELM by considering the structural relationship between low-dimension data and high-dimension data. As a kind of single-hidden layer feed-forward neural network (SLFN), ELM has the characteristics of fast training and good generalization ability compared to RBF. Affinity Propagation (AP) clustering algorithm can automatically determine the number of clusters. In this paper, a novel multi-label learning method named ML-AP-RBF-Lap-ELM is proposed which integrates AP clustering algorithm, ML-RBF and Lap-ELM. In this new model, the ML-RBF is used to map in the input layer. The number of hidden nodes and the center of the RBF function can be automatically determined by the AP clustering algorithm. The weights from the hidden layer to the output layer are solved by Lap-ELM. The simulation results show that ML-AP-RBF-Lap-ELM performs well on the three common data sets, including Natural Scene, Yeast Gene and 20NG (20 New Groups).

A novel partition model of scheduling regions for public bicycle system

Scheduling optimization, especially public bicycle scheduling analysis, has become a hot topic due to its benefits in social, economic and environmental aspects. The traditional methods mainly focus on the distance among bicycle stations to realize a self-balancing bicycle supply by minimizing the scheduling workload for each region but ignore the origin/destination flow-data and topological features of bicycle station networks. This paper proposes a novel partition model for public bicycle scheduling region. Besides the distance, the proposed model combines dynamic origin/destination flows of bicycle stations with the topological features of them in bicycle networks to construct a similarity matrix. Then scheduling region partition is obtained by using the affinity propagation clustering algorithm. To evaluate the performance of the proposed model, we compare it with the other baseline methods by calculating the correlation coefficient between the regions while the coupling coefficient functions on the real data from Hangzhou public bicycle system. The experimental result shows that the proposed model can achieve the smallest scheduling workload among the scheduling regions.

A Novel Extension to Fuzzy Connectivity for Body Composition Analysis: Applications in Thigh, Brain, and Whole Body Tissue Segmentation.

Magnetic resonance imaging (MRI) is the non-invasive modality of choice for body tissue composition analysis due to its excellent soft-tissue contrast and lack of ionizing radiation. However, quantification of body composition requires an accurate segmentation of fat, muscle, and other tissues from MR images, which remains a challenging goal due to the intensity overlap between them. In this study, we propose a fully automated, data-driven image segmentation platform that addresses multiple difficulties in segmenting MR images such as varying inhomogeneity, non-standardness, and noise, while producing a high-quality definition of different tissues. In contrast to most approaches in the literature, we perform segmentation operation by combining three different MRI contrasts and a novel segmentation tool, which takes into account variability in the data. The proposed system, based on a novel affinity definition within the fuzzy connectivity image segmentation family, prevents the need for user intervention and reparametrization of the segmentation algorithms. In order to make the whole system fully automated, we adapt an affinity propagation clustering algorithm to roughly identify tissue regions and image background. We perform a thorough evaluation of the proposed algorithm's individual steps as well as comparison with several approaches from the literature for the main application of muscle/fat separation. Furthermore, whole-body tissue composition and brain tissue delineation were conducted to show the generalization ability of the proposed system. This new automated platform outperforms other state-of-the-art segmentation approaches both in accuracy and efficiency.

An Adaptive Weighted KNN Positioning Method Based on Omnidirectional Fingerprint Database and Twice Affinity Propagation Clustering.

The human body has a great influence on Wi-Fi signal power. A fixed K value leads to localization errors for the K-nearest neighbor (KNN) algorithm. To address these problems, we present an adaptive weighted KNN positioning method based on an omnidirectional fingerprint database (ODFD) and twice affinity propagation clustering. Firstly, an OFPD is proposed to alleviate body’s sheltering impact on signal, which includes position, orientation and the sequence of mean received signal strength (RSS) at each reference point (RP). Secondly, affinity propagation clustering (APC) algorithm is introduced on the offline stage based on the fusion of signal-domain distance and position-domain distance. Finally, adaptive weighted KNN algorithm based on APC is proposed for estimating user’s position during online stage. K initial RPs can be obtained by KNN, then they are clustered by APC algorithm based on their position-domain distances. The most probable sub-cluster is reserved by the comparison of RPs’ number and signal-domain distance between sub-cluster center and the online RSS readings. The weighted average coordinates in the remaining sub-cluster can be estimated. We have implemented the proposed method with the mean error of 2.2 m, the root mean square error of 1.5 m. Experimental results show that our proposed method outperforms traditional fingerprinting methods.

Gravity Theory-Based Affinity Propagation Clustering Algorithm and Its Applications

Gravity Theory-Based Affinity Propagation Clustering Algorithm and Its Applications

Histogram Publishing Method Based on Differential Privacy

Differential privacy does not care about the background knowledge of attackers, and can strongly protect the data information to be released. The distribution of differential privacy histograms based on groupings has drawn much attention from researchers and how to balance the approximation error caused by the group mean with the Laplace error caused by adding noise is the key. This paper proposes a method of APG (Affinity Propagation Clustering and Grouping algorithm) based on clustering grouping to distribute differential privacy histogram, which can effectively balance the approximate error with the Laplace error and improve the histogram posting accuracy. In this method, the index mechanism is used firstly to realize the sorting of the buckets. Then, through the clustering of AP algorithms, the sorted buckets are adaptively grouped to find the optimal grouping strategy and release the data. Experimental results on real data show that APG method is superior to GS, AHP and IHP methods in accuracy of distribution.

Combined deep belief network in deep learning with affinity propagation clustering algorithm for roller bearings fault diagnosis without data label

Unlike many traditional feature extraction methods of vibration signal such as ensemble empirical mode decomposition (EEMD), deep belief network (DBN) in deep learning can extract the useful information automatically and reduce the reliance on experts, with signal processing technology, and troubleshooting experience. In conventional fault diagnosis, data labels are required for classifiers such as support vector machine, random forest, and artificial neural networks. These are usually based on expert knowledge, for training and testing. But the process is usually tedious. The clustering model, on the other hand, can finish the roller bearings fault diagnosis without data labels, which is more efficient. There are some common clustering models which include fuzzy C-means (FCM), Gustafson–Kessel (GK), Gath–Geva (GG) models, and affinity propagation (AP). Unlike FCM, GK, and GG, which require knowledge or experience to pre-set the number of cluster center points, AP clustering algorithm can obtain the cluster center point according to the responsibility and availability calculations for all data points automatically. To the best of the authors’ knowledge, AP is rarely used for fault diagnosis. In this paper, a method which combines DBN, with several hidden layers, and AP for roller bearings fault diagnosis is proposed. For data visualization, the principal component analysis (PCA) is deployed to reduce the dimension of the extracted feature. The first two principal components are employed as the input of the FCM, GK, GG, and AP models for roller bearings faults diagnosis. Compared with other combination models such as EEMD–PCA–FCM/GK/GG and DBN–PCA–FCM/GK/GG, the proposed method, from the experimental results, is superior to the aforementioned combination models.

A data-driven framework of typical treatment process extraction and evaluation

BackgroundA clinical pathway (CP) defines a standardized care process for a well-defined patient group that aims to improve patient outcomes and promote patient safety. However, the construction of a new pathway from scratch is a time-consuming task for medical staff because it involves many factors, including objects, multidisciplinary collaboration, sequential design, and outcome measurements. Recently, the rapid development of hospital information systems has allowed the storage of large volumes of electronic medical records (EMRs), and this information constitutes an abundant data resource for building CPs using data-mining methods. MethodsWe provide an automatic method for extracting typical treatment processes from EMRs that consists of four key steps. First, a novel similarity method is proposed to measure the similarity of two treatment records. Then, we perform an affinity propagation (AP) clustering algorithm to cluster doctor order set sequences (DOSSs). Next, a framework is proposed to extract a high-level description of each treatment cluster. Finally, we evaluate the extracted typical treatment processes by matching the treatment cluster with external information, such as the treatment efficacy, length of stay, and treatment cost. ResultsBy experiments on EMRs of 8287 cerebral infarction patients, it is concluded that our proposed method can effectively extract typical treatment processes from treatment records, and also has great potential to improve treatment outcome by personalizing the treatment process for patients with different conditions. ConclusionThe extracted typical treatment processes are intuitive and can provide managerial guidance for CP redesign and optimization. In addition, our work can assist clinicians in clearly understanding their routine treatment processes and recommend optimal treatment pathways for patients.

Affinity propagation clustering algorithm based on large-scale data-set

Affinity Propagation (AP) algorithm is not effective in processing large-scale data-sets, so the paper purposed an affinity propagation clustering algorithm based on large scale data-set, called LD-AP. First, we use the idea of grid clustering to divide large data-sets into small datasets and running AP in them to ensure the center of clustering. Then, we introduced the structure similarity matrix to calculate the distance of the cluster center. At last, we used Density peak Clustering Algorithm (DP) algorithm to cluster the cluster again. The experimental results show that the improved algorithm is better than the original algorithm in the clustering effect and computation speed.

Rank-Adaptive Non-Negative Matrix Factorization

Dimension reduction is a challenge task in data processing, especially in high-dimensional data processing area. Non-negative matrix factorization (NMF), as a classical dimension reduction method, has a contribution to the parts-based representation for the characteristics of non-negative constraints in the NMF algorithm. In this paper, the NMF algorithm is introduced to extract local features for dimension reduction. Considering the problem of which NMF is required to define the number of the decomposition rank manually, we proposed a rank-adaptive NMF algorithm, in which the affinity propagation (AP) clustering algorithm is introduced to determine adaptively the number of the decomposition rank of NMF. Then, the rank-adaptive NMF algorithm is used to extract features for the original image. After that, a low-dimensional representation of the original image is obtained through the projection from the original images to the feature space. Finally, we used extreme learning machine (ELM) and k-nearest neighbor (KNN) as the classifier to classify those low-dimensional feature representations. The experimental results demonstrate that the decomposition rank determined by the AP clustering algorithm can reflect the characteristics of the original data. When it is combined with the classification algorithm ELM or KNN and applied to handwritten character recognition, the proposed method not only reduces the dimension of original images but also performs well in terms of classification accuracy and time consumption. A new rank-adaptive NMF algorithm is proposed based on the AP clustering algorithm and the original NMF algorithm. According to this algorithm, the low-dimensional representation of the original data can be obtained without any prior knowledge. In addition, the proposed rank-adaptive NMF algorithm combined with the ELM and KNN classification algorithms performs well.

Recognition of genetically modified product based on affinity propagation clustering and terahertz spectroscopy

In this paper, based on the terahertz spectrum, a new identification method of genetically modified material by support vector machine (SVM) based on affinity propagation clustering is proposed. This algorithm mainly uses affinity propagation clustering algorithm to make cluster analysis and labeling on unlabeled training samples, and in the iterative process, the existing SVM training data are continuously updated, when establishing the identification model, it does not need to manually label the training samples, thus, the error caused by the human labeled samples is reduced, and the identification accuracy of the model is greatly improved.

Deployment Strategy for Car-Sharing Depots by Clustering Urban Traffic Big Data Based on Affinity Propagation

Car sharing is a type of car rental service, by which consumers rent cars for short periods of time, often charged by hours. The analysis of urban traffic big data is full of importance and significance to determine locations of depots for car-sharing system. Taxi OD (Origin-Destination) is a typical dataset of urban traffic. The volume of the data is extremely large so that traditional data processing applications do not work well. In this paper, an optimization method to determine the depot locations by clustering taxi OD points with AP (Affinity Propagation) clustering algorithm has been presented. By analyzing the characteristics of AP clustering algorithm, AP clustering has been optimized hierarchically based on administrative region segmentation. Considering sparse similarity matrix of taxi OD points, the input parameters of AP clustering have been adapted. In the case study, we choose the OD pairs information from Beijing’s taxi GPS trajectory data. The number and locations of depots are determined by clustering the OD points based on the optimization AP clustering. We describe experimental results of our approach and compare it with standard K-means method using quantitative and stationarity index. Experiments on the real datasets show that the proposed method for determining car-sharing depots has a superior performance.

Evolution Consistency Based Decomposition for Cooperative Coevolution

Cooperative coevolution has been proven a promising framework for large-scale optimization. However, its performance heavily relies on the problem decomposition strategy which decomposes a high dimensional problem into exclusive smaller sub-problems. Though many decomposition methods have been developed in recent years, they are confronted with limitations in capturing the interdependency among variables and costing a large number of fitness evaluations in the decomposition stage. To alleviate these issues, this paper proposes a data-driven decomposition method, which is called affinity propagation assisted and evolution consistency based decomposition, for cooperative coevolution. Specifically, we take advantage of historical evolutionary data to mine the evolution consistency among variables. Then, based on the mined consistency, we leverage the affinity propagation clustering algorithm to adaptively separate variables into groups with each group as a sub-problem. Particularly, this decomposition method is a dynamic variable grouping strategy, which is executed periodically during the evolution. The most advantageous property of this method is that it does not cost any fitness evaluations in the decomposition stage and could self-adaptively divide variables into groups. Extensive comparison results on two widely used large-scale benchmark sets demonstrate that the proposed decomposition method could assist the cooperative coevolution algorithm to achieve competitive or even better optimization performance than state-of-the-art decomposition methods. Therefore, the proposed decomposition strategy provides a new way to decompose variables into groups.

Hierarchical Prediction Based on Two-Level Affinity Propagation Clustering for Bike-Sharing System

Bike-sharing system is a new transportation that has emerged in recent years. More and more people will choose to ride bicycle sharing at home and abroad. While we use shared bicycles conveniently, there are also unfavorable factors that affect the customer’s riding experience in the bicycle-sharing system. Due to the rents or returns of bikes at different stations in different periods are imbalanced, the bikes in the system need to be rebalanced frequently. Therefore, there is an urgent need to predict and reallocate the bikes in advance. In this paper, we propose a hierarchical forecasting model that predicts the number of rents or returns to each station cluster in a future period to achieve redistribution. First, we propose a two-level affinity propagation clustering algorithm to divide bike stations into groups where migration trends of bikes among stations as well as geographical locations information are considered. Based on the two-level hierarchy of stations, the total rents of bikes are predicted. Then, we use a multi-similarity-based inference model to forecast the migration proportion of inter-cluster and across cluster, based on which the rents or returns of bikes at each station can be deduced. In order to verify the effectiveness of our two-level hierarchical prediction model, we validate it on the bike-sharing system of New York City and compare the results with those of other popular methods obtained. Experimental results demonstrate the superiority over other methods.

An Affinity Propagation Clustering Method Using Hybrid Kernel Function With LLE

Cluster analysis is important in data mining and clustering algorithms and has gained much attention during the last decade. However, it is a challenge to extract significant features from high-dimensional data and to rapidly provide satisfactory clustering results. This paper presents a new affinity propagation (AP) clustering method based on a hybrid kernel function with locally linear embedding, called LLE-HKAP, for the classification of gene expression datasets and standard UCI datasets. First, the locally linear embedding algorithm is used to reduce the dimension of the original dataset. Then, a novel AP clustering method based on a similarity measure with the hybrid kernel function is proposed. In this method, a new global kernel is defined that has high generalization ability. Meanwhile, a hybrid kernel function that linearly combines the proposed global kernel and the Gaussian kernel is defined to further enhance the learning ability of the global kernel. Moreover, the novel hybrid kernel is introduced to define a similarity measure and construct a similarity matrix of the AP clustering. Finally, the improved AP clustering algorithm is implemented on eight public gene expression datasets and eight standard UCI datasets for comparison with other related algorithms. The experimental results validate that our proposed clustering algorithm is efficient in terms of clustering accuracy and outperforms the currently available approaches with which it is compared.

Data-driven process decomposition and robust online distributed modelling for large-scale processes

ABSTRACTWith the increasing attention of networked control, system decomposition and distributed models show significant importance in the implementation of model-based control strategy. In this paper, a data-driven system decomposition and online distributed subsystem modelling algorithm was proposed for large-scale chemical processes. The key controlled variables are first partitioned by affinity propagation clustering algorithm into several clusters. Each cluster can be regarded as a subsystem. Then the inputs of each subsystem are selected by offline canonical correlation analysis between all process variables and its controlled variables. Process decomposition is then realised after the screening of input and output variables. When the system decomposition is finished, the online subsystem modelling can be carried out by recursively block-wise renewing the samples. The proposed algorithm was applied in the Tennessee Eastman process and the validity was verified.

Fault Detection Based on AP Clustering and PCA

To improve the accuracy, reduce the time consumption and obtain the number of faults, a fault detection method based on AP (affinity propagation) clustering and PCA (principal component analysis) was proposed. Firstly, discontinuous points in seismic horizons were searched out by the connected component labeling method. Secondly, the AP clustering algorithm was used to cluster the discontinuous points and the points of the same cluster were used to determine a fault, meanwhile, the faults existing in a seismic section were quantified. Finally, the PCA was adopted to calculate the principal direction of the discontinuous points contained in the same cluster. As a result, the corresponding cluster center and the principal direction determined a straight line, and the part that intercepted by the clustered edge was the fault we wanted. In the proposed method, the time consumption of correlation calculation of the traditional method was reduced; the computing work was simplified and the number of the faults in the seismic section was obtained. To confirm the feasibility and advancement of the proposed method, comparative experiments were done on the seismic model data and the real seismic section. The results show that the accuracy of the proposed method was better and the time cost was greatly reduced.

Optimized Automatic Seeded Region Growing Algorithm with Application to ROI Extraction

Region of interest (ROI) is the most important part of an image that expresses the effective content of the image. Extracting regions of interest from images accurately and efficiently can reduce computational complexity and is essential for image analysis and understanding. In order to achieve the automatic extraction of regions of interest and obtain more accurate regions of interest, this paper proposes Optimized Automatic Seeded Region Growing (OASRG) algorithm. The algorithm uses the affinity propagation (AP) clustering algorithm to extract the seeds automatically, and optimizes the traditional region growing algorithm by regrowing strategy to obtain the regions of interest where target objects are contained. Experimental results show that our algorithm can automatically locate seeds and produce results as good as traditional region growing with seeds selected manually. Furthermore, the precision is improved and the extraction effect is better after the optimization with regrowing strategy.