Density-based Clustering Approach Research Articles

DPC-DNG: Graph-based label propagation of [formula omitted]-nearest higher-density neighbors for density peaks clustering

Density peaks clustering (DPC) algorithm is a novel density-based clustering approach that effectively determines cluster centers from a decision graph and groups objects by assigning non-center objects to the same cluster as their nearest higher-density neighbor. Although DPC can allocate clusters of arbitrary shapes, its single-chain label propagation mechanism has the risk of “chain error”, where an object assigned an incorrect label causes its subordinates to also be assigned the same incorrect label. Hence, DPC is unable to effectively group objects that are located in overlapping areas between clusters, which leads to unsatisfactory clustering results. To address this issue, this study proposes the graph-based label propagation of k-nearest higher-density neighbor for density peaks clustering (DPC-DNG). DPC-DNG extends the single-chain label propagation of DPC to a graph-based multi-chain label propagation that assigns labels to objects from their k-nearest higher-density neighbors. First, based on k-nearest higher-density neighbors and the selected cluster centers, a symmetric density-based neighbor graph (DNG) is constructed. Second, to assign labels to objects, a classic graph-based label propagation mechanism is utilized in conjunction with DNG. To validate our method, we carry out comprehensive experiments on 6 synthetic and 12 real datasets. Statistically speaking, the results show that our method has improved the clustering performance of DPC and exhibits promising performance over other state-of-the-art DPC-related methods.

A comprehensive evaluation of OPTICS, GMM and K-means clustering methodologies for geochemical anomaly detection connected with sample catchment basins

The process of data-driven clustering to uncover geochemical anomalies linked to sample catchment basins (SCBs) includes a comprehensive framework to discern areas exhibiting unique geochemical attributes within a specified study area. The Ordering Points to Identify the Clustering Structure (OPTICS) method can serve as a robust methodology for detecting geochemical anomalies in SCBs. This is attributed to its capacity to effectively manage varying cluster densities, adaptively identify cluster numbers, exhibit resilience to noise, and display minimum sensitivity to parameters. A comparison was conducted in this research between the outcomes of the OPTICS clustering algorithm and two traditional clustering techniques, namely the Gaussian Mixture Model (GMM) and K-means clustering. In the following, the Expectation-Maximization (EM) technique is employed to train the GMM for clustering. Moreover, the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) as two validate statistical metrics implemented to ascertain the optimal number of components (clusters) belong to the GMM. It should be noted that the effectiveness of the clustering algorithms was further assessed using the Calinski-Harabasz (CH) index and the success-rate curves. OPTICS, a density-based clustering approach, was confirmed to be more effective than K-means and GMM for identifying MVT PbZn anomalies in Varcheh district, western Iran. Furthermore, the specified anomalies show a geo-spatial correspondence with the geological facts, and it has been observed that strong anomalies are more discoverable in close proximity to MVT PbZn occurrences. This work suggests a novel anomaly detection approach based on OPTICS, which exhibits superior performance and data-modeling efficiency. The main emphasis is on effectively distinguishing geochemical anomalies from sample data originating from populations with uncertain distributions.

Procedure code overutilization detection from healthcare claims using unsupervised deep learning methods

BackgroundFraud, Waste, and Abuse (FWA) in medical claims have a negative impact on the quality and cost of healthcare. A major component of FWA in claims is procedure code overutilization, where one or more prescribed procedures may not be relevant to a given diagnosis and patient profile, resulting in unnecessary and unwarranted treatments and medical payments. This study aims to identify such unwarranted procedures from millions of healthcare claims. In the absence of labeled examples of unwarranted procedures, the study focused on the application of unsupervised machine learning techniques.MethodsExperiments were conducted with deep autoencoders to find claims containing anomalous procedure codes indicative of FWA, and were compared against a baseline density-based clustering model. Diagnoses, procedures, and demographic data associated with healthcare claims were used as features for the models. A dataset of one hundred thousand claims sampled from a larger claims database is used to initially train and tune the models, followed by experimentations on a dataset with thirty-three million claims. Experimental results show that the autoencoder model, when trained with a novel feature-weighted loss function, outperforms the density-based clustering approach in finding potential outlier procedure codes.ResultsGiven the unsupervised nature of our experiments, model performance was evaluated using a synthetic outlier test dataset, and a manually annotated outlier test dataset. Precision, recall and F1-scores on the synthetic outlier test dataset for the autoencoder model trained on one hundred thousand claims were 0.87, 1.0 and 0.93, respectively, while the results for these metrics on the manually annotated outlier test dataset were 0.36, 0.86 and 0.51, respectively. The model performance on the manually annotated outlier test dataset improved further when trained on the larger thirty-three million claims dataset with precision, recall and F1-scores of 0.48, 0.90 and 0.63, respectively.ConclusionsThis study demonstrates the feasibility of leveraging unsupervised, deep-learning methods to identify potential procedure overutilization from healthcare claims.

A density connection weight-based clustering approach for dataset with density-sparse region

Density-Based clustering, an important clustering analysis approach, has several advantages of discovering clusters with arbitrary shapes and identifying noise samples etc. However, existing density-based clustering approaches cannot effectively identify clusters in density-sparse region. In this paper, we propose a density-based clustering approach for dataset with density-sparse region by using density connection weight. Firstly, we define a series of new concepts, such as density connection weight and average weight. And data objects are divided into core objects, boundary objects and noise objects according to average weight, which can effectively distinguish core objects and boundary objects in density-sparse region. Secondly, a density clustering assignment strategy is constructed based on density connection weight. A density clustering algorithm, together with the dimensionality reduction using t-SNE, judiciously handles the low-density clusters in multi-density datasets and the minimal clusters in extremely imbalanced datasets. In the end, the experimental results on the UCI and synthetic datasets validate that our algorithm exhibits prominent clustering performance, especially while being adapted in diverse clustering tasks, including the multi-density, imbalanced and uniform distribution datasets.

A Centralized Control-Based Clustering Scheme for Energy Efficiency in Underwater Acoustic Sensor Networks

Underwater acoustic sensor networks (UASNs) are among the most critical tools for underwater environment sensing and information acquisition. Since the complexity of the underwater environment renders battery replenishment and replacement more difficult and UASNs are mainly powered by batteries built into the underwater acoustic sensor nodes, the optimization of node energy consumption is particularly important for extending the network lifetime. Recently, energy consumption optimization solutions for UASNs have mainly focused on node clustering and multihop transmission. However, most schemes are implemented using distributed algorithms; in these types of approaches, decision information usually comes from only a subset of nodes, and it is difficult to make the most globally beneficial decision. Therefore, in this paper, the idea of centralized control is applied for the optimization of network energy consumption from a global perspective. Specifically, a centralized control-based clustering scheme (CCCS) for UASNs is proposed. In this scheme, a node density-based adaptive clustering approach is adopted, intracluster controllers are established within clusters, and the choices of both relay nodes and relay clusters are optimized to achieve energy balancing and route optimization. The performance achieved in simulations shows that compared to similar methods, CCCS is able to balance the node energy consumption and effectively increase the lifetime of a UASN.

A Robust Target Tracking Method for Crowded Indoor Environments Using mmWave Radar

Millimeter-wave-based extended target tracking has attracted extensive interest recently because of its privacy, high precision, and low cost. This paper concentrated on crowded indoor situations and presents a novel method for group tracking. First, the proposed alpha-extended Kalman filter and the group association were carried out, which can constantly estimate the target expansion and the number of reflection points, consequently modifying the measurement noise and covariance estimation. Then, to initialize the actual targets, we employed a density-based spatial clustering approach that includes false target suppression. After the targets have been updated, the track re-association and estimation procedure is conducted to account for the unanticipated break of moving and near-static targets. Finally, various experiments involving fewer than 11 participants were designed to assess the robustness of the method. As a result, continuous and steady tracking results, as well as high counting accuracy were obtained.

A data analytics framework for anomaly detection in flight operations

In the air transport system, there has been a continuous effort to develop policies, tools, and methodologies that increase and standardize safety levels across the entire commercial aviation market, while also enhancing operational efficiency. Furthermore, there is a current focus on proactive approaches for aviation performance management. Within this context, data mining initiatives such as anomaly detection have become more prominent. Dealing with anomalies is a natural step for achieving the goals regarding operational safety and efficiency, as anomalies are often related to hazardous and inefficient operations. In this work, we propose a systematic flight data analytics framework for anomaly detection in flight operations in order to provide a comprehensive and reusable pipeline for model building, application, and explanation. The solution is designed to be applicable to both online and offline regimes and at multiple scales, while also building on domain-expert analysis. We demonstrate the framework applicability in two scenarios of routine flight operations monitoring considering both airline and air traffic management perspectives. In the first one, the framework is applied to aircraft performance data within an unsupervised learning setting with a density-based clustering approach for anomaly detection in landing operations at Minneapolis–Saint Paul International Airport (KMSP). The results are compared with those obtained with exceedance-based methods used in the current practice, revealing the detection of operationally significant anomalies beyond the benchmark. In the second case study, we apply the framework on flight tracking data within a supervised learning setting with the development of an autoencoder classifier for offline anomaly detection in terminal airspace arrival operations at Sao Paulo/Guarulhos International Airport (SBGR). Additionally, supervised learning models are developed for anomaly explanation. The autoencoder classifier was able to detect operationally significant anomalies, while the explanatory models provided novel insights about contributing factors to the anomalies identified. For instance, we learned that anomalous arrival trajectories are more likely to be associated with landing operations on runway 27 under wind scenarios, with an increase in the odds ratio of 62% and 58% for tailwinds and headwinds, respectively. In addition, we also observed a positive association between anomalies and wind gusts situations.

Fast retrieval and efficient identification of monument images using features based adaptive clustering and optimized deep belief network

ABSTRACT Content-Based Image Retrieval (CBIR) is an extensively used image processing technique for retrieving similar images (general, images, monument images, etc) from large datasets. Some existing approaches' accuracy and retrieval time are ineffective for monument image retrieval. To overcome these challenges, the monument images are identified and retrieved efficiently using the visual feature extraction and adaptive density-based clustering approach (ADCA). Features required for efficient retrieval is obtained and the obtained multiple features are fused into a single feature using the hybrid weighted and average weighted methods. An optimized Deep belief network (ODBN) is designed to label the clusters based on the grouped images and then retrieve the images relevant to query images. From the experimental results, the overall accuracy achieved by the proposed framework are 99.63% (Architectural heritage dataset), 98.88% (synthetic dataset) and 99.1% (Kaggle dataset), respectively. The experimental results proved the efficiency of proposed in retrieving monument images.

Efficient Trajectory Clustering with Road Network Constraints Based on Spatiotemporal Buffering

The analysis of individuals’ movement behaviors is an important area of research in geographic information sciences, with broad applications in smart mobility and transportation systems. Recent advances in information and communication technologies have enabled the collection of vast amounts of mobility data for investigating movement behaviors using trajectory data mining techniques. Trajectory clustering is one commonly used method, but most existing methods require a complete similarity matrix to quantify the similarities among users’ trajectories in the dataset. This creates a significant computational overhead for large datasets with many user trajectories. To address this complexity, an efficient clustering-based method for network constraint trajectories is proposed, which can help with transportation planning and reduce traffic congestion on roads. The proposed algorithm is based on spatiotemporal buffering and overlapping operations and involves the following steps: (i) Trajectory preprocessing, which uses an efficient map-matching algorithm to match trajectory points to the road network. (ii) Trajectory segmentation, where a Compressed Linear Reference (CLR) technique is used to convert the discrete 3D trajectories to 2D CLR space. (iii) Spatiotemporal proximity analysis, which calculates a partial similarity matrix using the Longest Common Subsequence similarity indicator in CLR space. (iv) Trajectory clustering, which uses density-based and hierarchical clustering approaches to cluster the trajectories. To verify the proposed clustering-based method, a case study is carried out using real trajectories from the GeoLife project of Microsoft Research Asia. The case study results demonstrate the effectiveness and efficiency of the proposed method compared with other state-of-the-art clustering-based methods.

A Secure Data Sharing Scheme in Community Segmented Vehicular Social Networks for 6G

The use of aerial base stations, AI cloud, and satellite storage can help manage location, traffic, and specific application-based services for vehicular social networks. However, sharing of such data makes the vehicular network vulnerable to data and privacy leakage. In this regard, this article proposes an efficient and secure data sharing scheme using community segmentation and a blockchain-based framework for vehicular social networks. The proposed work considers similarity matrices that employ the dynamics of structural similarity, modularity matrix, and data compatibility. These similarity matrices are then passed through stacked autoencoders that are trained to extract encoded embedding. A density-based clustering approach is then employed to find the community segments from the information distances between the encoded embeddings. A blockchain network based on the Hyperledger Fabric platform is also adopted to ensure data sharing security. Extensive experiments have been carried out to evaluate the proposed data-sharing framework in terms of the sum of squared error, sharing degree, time cost, computational complexity, throughput, and CPU utilization for proving its efficacy and applicability. The results show that the CSB framework achieves a higher degree of SD, lower computational complexity, and higher throughput.

Subspace Clustering in High-Dimensional Data Streams: A Systematic Literature Review

Clustering high dimensional data is challenging as data dimensionality increases the distance between data points, resulting in sparse regions that degrade clustering performance. Subspace clustering is a common approach for processing high-dimensional data by finding relevant features for each cluster in the data space. Subspace clustering methods extend traditional clustering to account for the constraints imposed by data streams. Data streams are not only high-dimensional, but also unbounded and evolving. This necessitates the development of subspace clustering algorithms that can handle high dimensionality and adapt to the unique characteristics of data streams. Although many articles have contributed to the literature review on data stream clustering, there is currently no specific review on subspace clustering algorithms in high-dimensional data streams. Therefore, this article aims to systematically review the existing literature on subspace clustering of data streams in high-dimensional streaming environments. The review follows a systematic methodological approach and includes 18 articles for the final analysis. The analysis focused on two research questions related to the general clustering process and dealing with the unbounded and evolving characteristics of data streams. The main findings relate to six elements: clustering process, cluster search, subspace search, synopsis structure, cluster maintenance, and evaluation measures. Most algorithms use a two-phase clustering approach consisting of an initialization stage, a refinement stage, a cluster maintenance stage, and a final clustering stage. The density-based top-down subspace clustering approach is more widely used than the others because it is able to distinguish true clusters and outliers using projected micro-clusters. Most algorithms implicitly adapt to the evolving nature of the data stream by using a time fading function that is sensitive to outliers. Future work can focus on the clustering framework, parameter optimization, subspace search techniques, memory-efficient synopsis structures, explicit cluster change detection, and intrinsic performance metrics. This article can serve as a guide for researchers interested in high-dimensional subspace clustering methods for data streams.

DPC-FSC: An approach of fuzzy semantic cells to density peaks clustering

Density peaks clustering (DPC) algorithm is a succinct and efficient density-based clustering approach to data analysis. It computes the local density and the relative distance for objects to seek cluster centers and form clusters. However, it is difficult to estimate an appropriate local density by a rule of thumb; therefore, the performance of DPC may be poor on real datasets in practice. Thus, this study proposes a novel method for density peaks clustering based on fuzzy semantic cells. Specifically, each object is coarsened into a fuzzy point with the form of the fuzzy semantic cell model. Based on this model, a local density metric is defined and estimated via the principle of justifiable granularity. Our local density estimation can be converted into an optimization problem. In addition, a relative semantic distance is also introduced which concerns the distance between fuzzy semantic cells. The relative semantic distance is more informative for selecting cluster centers in the decision graph. The experimental results show that our method not only exhibits higher performance but also provides a clearer decision graph to select cluster centers.

An adaptive clustering algorithm by neighbourhood search for large-scale data

Adaptive cluster sampling (ACS) is a sampling method relies on the neighbourhood search on a grid structure. It has an adaptive selection process of units and recursively added units reveal the batched individuals easily and quickly. In this paper, we propose a new clustering method called spatial adaptive clustering (SAC) based on the idea of ACS design. The SAC algorithm forms clusters based on neighbourhood search using grid structures and is able to detect noise points. The performance of the proposed algorithm is evaluated through comparison with the results from well-known density-based clustering approaches in the literature using real and artificial data sets. Computational results indicate that the proposed algorithm is effective in terms of external validation measures for clustering of arbitrary shaped data with noise. Additionally, the SAC algorithm is tested on artificial data sets of varying sizes for the runtime criterion. The results reveal that it also performs superbly for the objective of reducing the runtime.

3D point cloud density-based segmentation for vine rows detection and localisation

The adoption of new sensors for crop monitoring is leading to the acquisition of large amounts of data, which usually are not directly usable for agricultural applications. The 3D point cloud maps of fields and parcels, generated from remotely sensed data, are examples of such big data, which require the development of specific algorithms for their processing and interpretation, with the final aim to obtain valuable information about crop status.This manuscript presents an innovative 3D point cloud processing algorithm for vine row detection and localisation within vineyard maps, based on the detection of key points and a density-based clustering approach. Vine row localisation is a crucial phase in the interpretation of the complex and huge 3D point clouds of agricultural environments, which makes it possible to move the focus from a macro level (parcel and plot scale) to a micro level (plants, fruits and branches). The algorithm outputs fully describe the spatial location of each vine row within the whole 3D model of the agricultural environment by a set of key points and an interpolating curve. The algorithm is specifically conceived to be robust and: (i) independent of the adopted airborne sensor used to acquire the in-field data (not requiring a model with colour or spectral information); (ii) able to manage vineyards with any vine row layout or orientation (such as curvilinear) and (iii) not hindered by the occurrence of missing plants. The experimental results, obtained by processing the models of seven case study parcels, proved the algorithm’s reliability and accuracy: the automatic vine row detection was found to be 100% in accordance with the manual one; and the obtained localisation indices showed an average error of 12 cm and standard deviation of 10 cm, which is fully compatible with the considered agricultural applications. In addition, the algorithm outputs can be profitably exploited for enhanced path planning of autonomous agricultural machines adopted for in-field operations.

Density-based clustering and probabilistic classification for integrated transmission-distribution network security state prediction

The proliferation of renewable energy sources (RESs) into the distribution network necessitates the need for the capability to predict the security state of the grid. This paper proposes a density-based clustering and probabilistic classification approach to predict the security state of the modern grid. Firstly, an approach to predict the critical clearing time using the changes in inertia constant and system load is proposed. Secondly, an algorithm for training dataset development from the transient stability responses of the grid considering different operation scenarios is proposed. An expectation-maximization (EM) algorithm using the density-based clustering technique was applied to the dataset to obtain clusters representing the network's security states. Finally, a predictive model was obtained from the labeled dataset using a Naïve-Bayes (NB) probabilistic classifier. The feasibility of this approach is demonstrated using the IEEE 14 bus system. Respectively, an APA and RMSE of 98% and 3.6% were obtained. Also, an MAE of less than 1% was obtained when the proposed model was tested with seven different datasets having a different number of instances. The results show that the proposed technique can predict the security of the integrated transmission-distribution networks considering different DG types, penetration levels, and network disturbances with a high degree of accuracy.

Accurate Sensing of Power Transformer Faults From Dissolved Gas Data Using Random Forest Classifier Aided by Data Clustering Method

In this paper, a novel approach for accurate sensing of incipient faults occurring in power transformers is proposed using dissolved gas analysis (DGA) technique. The Duval pentagon method is a popular technique often used to interpret faults occurring in a power transformer based on DGA data. However, one potential limitation of conventional Duval pentagon method is the presence of rigid fault boundaries within the pentagon which often lead to misinterpretations, leading to poor detection accuracy. Considering this issue, in this paper a modification of Duval pentagon method is proposed, where instead of using rigidly separated distinct fault zones, a density-based clustering (DBSCAN) approach is used to increase the resiliency and the accuracy of fault detection technique. At first, DBSCAN is used to form different fault clusters within the Duval pentagon. Following this, the centroid corresponding to each fault cluster within the Duval pentagon is determined. For accurate sensing of incipient transformer faults Euclidean distances between the respective centroids and the fault points of the input DGA data are proposed as new distinguishing features in this work. The proposed distance parameters combined with the relative gas concentration measures are finally served as input features to the random forest (RF) classifier, which returned very high classification accuracy. The performance of the RF classifier is also compared with three benchmark machine classifiers, all of which delivered acceptable results. The proposed method can be used for sensing of power transformer faults using Duval pentagon method with increased accuracy.

Density Peaks Clustering Based on Natural Search Neighbors and Manifold Distance Metric

The famous density-based clustering approach Density Peaks Clustering (DPC) is getting more and more popular recently. However, DPC algorithm is vulnerable to the parameter <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d_c</i> and is incapable of obtaining desired clustering results when dealing with manifold data sets. Furthermore, the allocation strategy of the remaining data points can easily lead to domino chain reaction which means that when a data point is allocated improperly, the points around it will be allocated incorrectly too. To solve these deficiencies, in this paper, we put forward an algorithm named density peaks clustering based on natural search neighbors and manifold distance metric (DPC-NSN-MDM). In the beginning, we apply natural search method (NSM) to identify the natural neighbors and further calculate the local density ρ of each data point. Secondly, when calculating distance δ of each data point, we put the manifold distance metric with a scaling factor to replace the traditional Euclidean distance metric. At last, manifold distance is also used in the allocation strategy of remaining data points to reduce the effect of the domino chain reaction. The proposed DPC-NSN-MDM algorithm succeeds in getting superb experimental performance on both synthetic data sets and real-world data sets.

Effective density-based clustering algorithms for incomplete data

Density-based clustering is an important category among clustering algorithms. In real applications, many datasets suffer from incompleteness. Traditional imputation technologies or other techniques for handling missing values are not suitable for density-based clustering and decrease clustering result quality. To avoid these problems, we develop a novel density-based clustering approach for incomplete data based on Bayesian theory, which conducts imputation and clustering concurrently and makes use of intermediate clustering results. To avoid the impact of low-density areas inside non-convex clusters, we introduce a local imputation clustering algorithm, which aims to impute points to high-density local areas. The performances of the proposed algorithms are evaluated using ten synthetic datasets and five real-world datasets with induced missing values. The experimental results show the effectiveness of the proposed algorithms.

SDCOR: Scalable density-based clustering for local outlier detection in massive-scale datasets

This paper presents a batch-wise density-based clustering approach for local outlier detection in massive-scale datasets. Unlike the well-known traditional algorithms, which assume that all the data is memory-resident, our proposed method is scalable and processes the input data chunk-by-chunk within the confines of a limited memory buffer. A temporary clustering model is built at the first phase; then, it is gradually updated by analyzing consecutive memory loads of points. Subsequently, at the end of scalable clustering, the approximate structure of the original clusters is obtained. Finally, by another scan of the entire dataset and using a suitable criterion, an outlying score is assigned to each object called SDCOR (Scalable Density-based Clustering Outlierness Ratio). Evaluations on real-life and synthetic datasets demonstrate that the proposed method has a low linear time complexity and is more effective and efficient compared to best-known conventional density-based methods, which need to load all data into the memory; and also, to some fast distance-based methods, which can perform on data resident in the disk.

A Novel Approach for Density-Based Optimal Semantic Clustering of Web Objects via Identification of KingPins

Background: Clustering is one of the important techniques in Data Mining to group the related data. Clustering can be applied on numerical data as well as web objects such as URLs, websites, documents, keywords etc. which is the building block for many recommender systems as well as prediction models. Objective: The objective of this research article is to develop an optimal clustering approach which considers semantics of web objects to cluster them in a group. More so importantly, the purpose of the proposed work is to strictly improve the computation time of clustering process. Methods: In order to achieve the desired objectives, following two contributions have been proposed to improve the clustering approach 1) Semantic Similarity Measure based on Wu-Palmer Semantics based similarity 2). Two-Level Densitybased Clustering technique to reduce the computational complexity of density based clustering approach. Results: The efficacy of the proposed method has been analyzed on AOL search logs containing 20 million web queries. The results showed that our approach increases the F-measure, and decreases the entropy. It also reduces the computational complexity and provides a competitive alternative strategy of semantic clustering when conventional methods do not provide helpful suggestions. Conclusion: A clustering model has been proposed which is composed of two components i.e. Similarity measure and Density based two-level clustering technique. The proposed model reduced the time cost of density based clustering approach without effecting the performance.