GNN-DBSCAN: A new density-based algorithm using grid and the nearest neighbor

2021 ◽  
pp. 1-13
Author(s):  
Li Yihong ◽  
Wang Yunpeng ◽  
Li Tao ◽  
Lan Xiaolong ◽  
Song Han
Keyword(s):  
Mutual Information ◽  
Clustering Algorithm ◽  
Nearest Neighbor ◽  
Spatial Clustering ◽  
Clustering Algorithms ◽  
Adjusted Rand Index ◽  
K Nearest Neighbors ◽  
Normalized Mutual Information ◽  
Core Samples ◽  
Real World Datasets

DBSCAN (density-based spatial clustering of applications with noise) is one of the most widely used density-based clustering algorithms, which can find arbitrary shapes of clusters, determine the number of clusters, and identify noise samples automatically. However, the performance of DBSCAN is significantly limited as it is quite sensitive to the parameters of eps and MinPts. Eps represents the eps-neighborhood and MinPts stands for a minimum number of points. Additionally, a dataset with large variations in densities will probably trap the DBSCAN because its parameters are fixed. In order to overcome these limitations, we propose a new density-clustering algorithm called GNN-DBSCAN which uses an adaptive Grid to divide the dataset and defines local core samples by using the Nearest Neighbor. With the help of grid, the dataset space will be divided into a finite number of cells. After that, the nearest neighbor lying in every filled cell and adjacent filled cells are defined as the local core samples. Then, GNN-DBSCAN obtains global core samples by enhancing and screening local core samples. In this way, our algorithm can identify higher-quality core samples than DBSCAN. Lastly, give these global core samples and use dynamic radius based on k-nearest neighbors to cluster the datasets. Dynamic radius can overcome the problems of DBSCAN caused by its fixed parameter eps. Therefore, our method can perform better on dataset with large variations in densities. Experiments on synthetic and real-world datasets were conducted. The results indicate that the average Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Adjusted Mutual Information (AMI) and V-measure of our proposed algorithm outperform the existing algorithm DBSCAN, DPC, ADBSCAN, and HDBSCAN.

scholarly journals DeepDBSCAN: Deep Density-Based Clustering for Geo-Tagged Photos

2021 ◽  
Vol 10 (8) ◽  
pp. 548
Author(s):  
Jang-You Park ◽  
Dong-June Ryu ◽  
Kwang-Woo Nam ◽  
Insung Jang ◽  
Minseok Jang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Clustering Algorithm ◽  
Nearest Neighbor ◽  
Spatial Clustering ◽  
Clustering Algorithms ◽  
Learning Technology ◽  
Neighbor Graph ◽  
Density Based Clustering ◽  
Real Objects ◽  
Nearest Neighbor Graph

Density-based clustering algorithms have been the most commonly used algorithms for discovering regions and points of interest in cities using global positioning system (GPS) information in geo-tagged photos. However, users sometimes find more specific areas of interest using real objects captured in pictures. Recent advances in deep learning technology make it possible to recognize these objects in photos. However, since deep learning detection is a very time-consuming task, simply combining deep learning detection with density-based clustering is very costly. In this paper, we propose a novel algorithm supporting deep content and density-based clustering, called deep density-based spatial clustering of applications with noise (DeepDBSCAN). DeepDBSCAN incorporates object detection by deep learning into the density clustering algorithm using the nearest neighbor graph technique. Additionally, this supports a graph-based reduction algorithm that reduces the number of deep detections. We performed experiments with pictures shared by users on Flickr and compared the performance of multiple algorithms to demonstrate the excellence of the proposed algorithm.

scholarly journals Clustering by Detecting Density Peaks and Assigning Points by Similarity-First Search Based on Weighted K-Nearest Neighbors Graph

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-17
Author(s):  
Qi Diao ◽  
Yaping Dai ◽  
Qichao An ◽  
Weixing Li ◽  
Xiaoxue Feng ◽  
...  
Keyword(s):  
Clustering Algorithm ◽  
Spatial Clustering ◽  
Local Density ◽  
Search Algorithm ◽  
Real Data ◽  
Nearest Neighbors ◽  
Adjusted Rand Index ◽  
Clustering Methods ◽  
K Nearest Neighbors ◽  
Density Peaks

This paper presents an improved clustering algorithm for categorizing data with arbitrary shapes. Most of the conventional clustering approaches work only with round-shaped clusters. This task can be accomplished by quickly searching and finding clustering methods for density peaks (DPC), but in some cases, it is limited by density peaks and allocation strategy. To overcome these limitations, two improvements are proposed in this paper. To describe the clustering center more comprehensively, the definitions of local density and relative distance are fused with multiple distances, including K-nearest neighbors (KNN) and shared-nearest neighbors (SNN). A similarity-first search algorithm is designed to search the most matching cluster centers for noncenter points in a weighted KNN graph. Extensive comparison with several existing DPC methods, e.g., traditional DPC algorithm, density-based spatial clustering of applications with noise (DBSCAN), affinity propagation (AP), FKNN-DPC, and K-means methods, has been carried out. Experiments based on synthetic data and real data show that the proposed clustering algorithm can outperform DPC, DBSCAN, AP, and K-means in terms of the clustering accuracy (ACC), the adjusted mutual information (AMI), and the adjusted Rand index (ARI).

DRSA: a non-hierarchical clustering algorithm using k-NN graph and its application in vegetation classification

2015 ◽  
pp. 125-138 ◽  
Author(s):  
I. V. Goncharenko
Keyword(s):  
Cluster Analysis ◽  
Clustering Algorithm ◽  
Nearest Neighbor ◽  
Clustering Algorithms ◽  
Protein Structures ◽  
Hierarchical Cluster ◽  
Vegetation Classification ◽  
K Nearest Neighbor ◽  
Neighbor Graph ◽  
Nearest Neighbor Graph

In this article we proposed a new method of non-hierarchical cluster analysis using k-nearest-neighbor graph and discussed it with respect to vegetation classification. The method of k-nearest neighbor (k-NN) classification was originally developed in 1951 (Fix, Hodges, 1951). Later a term “k-NN graph” and a few algorithms of k-NN clustering appeared (Cover, Hart, 1967; Brito et al., 1997). In biology k-NN is used in analysis of protein structures and genome sequences. Most of k-NN clustering algorithms build «excessive» graph firstly, so called hypergraph, and then truncate it to subgraphs, just partitioning and coarsening hypergraph. We developed other strategy, the “upward” clustering in forming (assembling consequentially) one cluster after the other. Until today graph-based cluster analysis has not been considered concerning classification of vegetation datasets.

scholarly journals An Enhanced Spectral Clustering Algorithm with S-Distance

Symmetry ◽  
2021 ◽  
Vol 13 (4) ◽  
pp. 596
Author(s):  
Krishna Kumar Sharma ◽  
Ayan Seal ◽  
Enrique Herrera-Viedma ◽  
Ondrej Krejcar
Keyword(s):  
Spectral Clustering ◽  
Clustering Algorithm ◽  
Spatial Clustering ◽  
Clustering Algorithms ◽  
Rank Test ◽  
Customer Churn ◽  
Signed Rank ◽  
Signed Rank Test ◽  
Spectral Clustering Algorithm ◽  
Industrial Databases

Calculating and monitoring customer churn metrics is important for companies to retain customers and earn more profit in business. In this study, a churn prediction framework is developed by modified spectral clustering (SC). However, the similarity measure plays an imperative role in clustering for predicting churn with better accuracy by analyzing industrial data. The linear Euclidean distance in the traditional SC is replaced by the non-linear S-distance (Sd). The Sd is deduced from the concept of S-divergence (SD). Several characteristics of Sd are discussed in this work. Assays are conducted to endorse the proposed clustering algorithm on four synthetics, eight UCI, two industrial databases and one telecommunications database related to customer churn. Three existing clustering algorithms—k-means, density-based spatial clustering of applications with noise and conventional SC—are also implemented on the above-mentioned 15 databases. The empirical outcomes show that the proposed clustering algorithm beats three existing clustering algorithms in terms of its Jaccard index, f-score, recall, precision and accuracy. Finally, we also test the significance of the clustering results by the Wilcoxon’s signed-rank test, Wilcoxon’s rank-sum test, and sign tests. The relative study shows that the outcomes of the proposed algorithm are interesting, especially in the case of clusters of arbitrary shape.

An improved OPTICS clustering algorithm for discovering clusters with uneven densities

2021 ◽  
Vol 25 (6) ◽  
pp. 1453-1471
Author(s):  
Chunhua Tang ◽  
Han Wang ◽  
Zhiwen Wang ◽  
Xiangkun Zeng ◽  
Huaran Yan ◽  
...  
Keyword(s):  
Time Complexity ◽  
Clustering Algorithm ◽  
Nearest Neighbor ◽  
Clustering Algorithms ◽  
Substantial Improvement ◽  
Experimental Results ◽  
High Time ◽  
Parameter Setting ◽  
K Nearest Neighbor ◽  
Density Based Clustering

Most density-based clustering algorithms have the problems of difficult parameter setting, high time complexity, poor noise recognition, and weak clustering for datasets with uneven density. To solve these problems, this paper proposes FOP-OPTICS algorithm (Finding of the Ordering Peaks Based on OPTICS), which is a substantial improvement of OPTICS (Ordering Points To Identify the Clustering Structure). The proposed algorithm finds the demarcation point (DP) from the Augmented Cluster-Ordering generated by OPTICS and uses the reachability-distance of DP as the radius of neighborhood eps of its corresponding cluster. It overcomes the weakness of most algorithms in clustering datasets with uneven densities. By computing the distance of the k-nearest neighbor of each point, it reduces the time complexity of OPTICS; by calculating density-mutation points within the clusters, it can efficiently recognize noise. The experimental results show that FOP-OPTICS has the lowest time complexity, and outperforms other algorithms in parameter setting and noise recognition.

scholarly journals An Improved Version of K-medoid Algorithm using CRO

2018 ◽  
Vol 12 (2) ◽  
pp. 116 ◽  
Author(s):  
Amjad Hudaib ◽  
Mohammad Khanafseh ◽  
Ola Surakhi
Keyword(s):  
Breast Cancer ◽  
Lung Cancer ◽  
Hybrid Algorithm ◽  
Clustering Algorithm ◽  
Clustering Algorithms ◽  
Data Repository ◽  
The Mean ◽  
Real World Datasets ◽  
Actual Point ◽  
Learning Data

Clustering is the process of grouping a set of patterns into different disjoint clusters where each cluster contains the alike patterns. Many algorithms had been proposed before for clustering. K-medoid is a variant of k-mean that use an actual point in the cluster to represent it instead of the mean in the k-mean algorithm to get the outliers and reduce noise in the cluster. In order to enhance performance of k-medoid algorithm and get more accurate clusters, a hybrid algorithm is proposed which use CRO algorithm along with k-medoid. In this method, CRO is used to expand searching for the optimal medoid and enhance clustering by getting more precise results. The performance of the new algorithm is evaluated by comparing its results with five clustering algorithms, k-mean, k-medoid, DB/rand/1/bin, CRO based clustering algorithm and hybrid CRO-k-mean by using four real world datasets: Lung cancer, Iris, Breast cancer Wisconsin and Haberman’s survival from UCI machine learning data repository. The results were conducted and compared base on different metrics and show that proposed algorithm enhanced clustering technique by giving more accurate results.

scholarly journals EVALUATION OF MUTUAL INFORMATION ESTIMATORS FOR TIME SERIES

2009 ◽  
Vol 19 (12) ◽  
pp. 4197-4215 ◽  
Author(s):  
ANGELIKI PAPANA ◽  
DIMITRIS KUGIUMTZIS
Keyword(s):  
Time Series ◽  
Nonlinear Systems ◽  
Mutual Information ◽  
Nearest Neighbor ◽  
K Nearest Neighbor ◽  
Series Length ◽  
K Nearest Neighbors ◽  
Data Adaptive ◽  
The Comparative Study ◽  
Selection Of

We study some of the most commonly used mutual information estimators, based on histograms of fixed or adaptive bin size, k-nearest neighbors and kernels and focus on optimal selection of their free parameters. We examine the consistency of the estimators (convergence to a stable value with the increase of time series length) and the degree of deviation among the estimators. The optimization of parameters is assessed by quantifying the deviation of the estimated mutual information from its true or asymptotic value as a function of the free parameter. Moreover, some commonly used criteria for parameter selection are evaluated for each estimator. The comparative study is based on Monte Carlo simulations on time series from several linear and nonlinear systems of different lengths and noise levels. The results show that the k-nearest neighbor is the most stable and less affected by the method-specific parameter. A data adaptive criterion for optimal binning is suggested for linear systems but it is found to be rather conservative for nonlinear systems. It turns out that the binning and kernel estimators give the least deviation in identifying the lag of the first minimum of mutual information from nonlinear systems, and are stable in the presence of noise.

scholarly journals Adaptive Initialization Method for K-Means Algorithm

2021 ◽  
Vol 4 ◽  
Author(s):  
Jie Yang ◽  
Yu-Kai Wang ◽  
Xin Yao ◽  
Chin-Teng Lin
Keyword(s):  
Time Complexity ◽  
Clustering Algorithm ◽  
Clustering Algorithms ◽  
Real Life ◽  
Superior Performance ◽  
Local Optima ◽  
Initial Cluster ◽  
Higher Dimensional ◽  
Real World Datasets ◽  
Random Method

The K-means algorithm is a widely used clustering algorithm that offers simplicity and efficiency. However, the traditional K-means algorithm uses a random method to determine the initial cluster centers, which make clustering results prone to local optima and then result in worse clustering performance. In this research, we propose an adaptive initialization method for the K-means algorithm (AIMK) which can adapt to the various characteristics in different datasets and obtain better clustering performance with stable results. For larger or higher-dimensional datasets, we even leverage random sampling in AIMK (name as AIMK-RS) to reduce the time complexity. 22 real-world datasets were applied for performance comparisons. The experimental results show AIMK and AIMK-RS outperform the current initialization methods and several well-known clustering algorithms. Specifically, AIMK-RS can significantly reduce the time complexity to O (n). Moreover, we exploit AIMK to initialize K-medoids and spectral clustering, and better performance is also explored. The above results demonstrate superior performance and good scalability by AIMK or AIMK-RS. In the future, we would like to apply AIMK to more partition-based clustering algorithms to solve real-life practical problems.

A novel dual-domain clustering algorithm for inhomogeneous spatial point event

2020 ◽  
Vol 54 (5) ◽  
pp. 603-623
Author(s):  
Jie Zhu ◽  
Jing Yang ◽  
Shaoning Di ◽  
Jiazhu Zheng ◽  
Leying Zhang
Keyword(s):  
Clustering Algorithm ◽  
Spatial Clustering ◽  
Clustering Algorithms ◽  
Spatial Domain ◽  
Spatial Proximity ◽  
Land Use Classification ◽  
Content Type ◽  
Spatial Point ◽  
Spatial Attributes ◽  
Dual Domain

PurposeThe spatial and non-spatial attributes are the two important characteristics of a spatial point, which belong to the two different attribute domains in many Geographic Information Systems applications. The dual clustering algorithms take into account both spatial and non-spatial attributes, where a cluster has not only high proximity in spatial domain but also high similarity in non-spatial domain. In a geographical dataset, traditional dual spatial clustering algorithms discover homogeneous spatially adjacent clusters suffering from the between-cluster inhomogeneity where those spatial points are described in non-spatial domain. To overcome this limitation, a novel dual-domain clustering algorithm (DDCA) is proposed by considering both spatial proximity and attribute similarity with the presence of inhomogeneity.Design/methodology/approachIn this algorithm, Delaunay triangulation with edge length constraints is first employed to construct spatial proximity relationships amongst objects. Then, a clustering strategy based on statistical change detection is designed to obtain clusters with similar attributes.FindingsThe effectiveness and practicability of the proposed algorithm are illustrated by experiments on both simulated datasets and real spatial events. It is found that the proposed algorithm can adaptively and accurately detect clusters with spatial proximity and similar non-spatial attributes under the consideration of inhomogeneity.Originality/valueTraditional dual spatial clustering algorithms discover homogeneous spatially adjacent clusters suffering from the between-cluster inhomogeneity where those spatial points are described in non-spatial domain. The research here is a contribution to developing a dual spatial clustering method considering both spatial proximity and attribute similarity with the presence of inhomogeneity. The detection of these clusters is useful to understand the local patterns of geographical phenomena, such as land use classification, spatial patterns research and big geo-data analysis.

Semi-Supervised Clustering Algorithm Based on Small Size of Labeled Data

2011 ◽  
Vol 121-126 ◽  
pp. 4675-4679
Author(s):  
Ming Wei Leng ◽  
Xiao Yun Chen ◽  
Jian Jun Cheng ◽  
Long Jie Li
Keyword(s):  
Data Mining ◽  
Clustering Algorithm ◽  
Clustering Algorithms ◽  
Nearest Neighbors ◽  
Experimental Results ◽  
K Nearest Neighbors ◽  
Supervised Clustering ◽  
The Core ◽  
Knn Classification ◽  
Core Problem

In many data mining domains, labeled data is very expensive to generate, how to make the best use of labeled data to guide the process of unlabeled clustering is the core problem of semi-supervised clustering. Most of semi-supervised clustering algorithms require a certain amount of labeled data and need set the values of some parameters, different values maybe have different results. In view of this, a new algorithm, called semi-supervised clustering algorithm based on small size of labeled data, is presented, which can use the small size of labeled data to expand labeled dataset by labeling their k-nearest neighbors and only one parameter. We demonstrate our clustering algorithm with three UCI datasets, compared with SSDBSCAN[4] and KNN, the experimental results confirm that accuracy of our clustering algorithm is close to that of KNN classification algorithm.

Sign in / Sign up

Export Citation Format

Share Document