Clustering Based on Density Propagation and Subcluster Merging

Clustering data streams has drawn lots of attention in the last few years due to their ever-growing presence. Data streams put additional challenges on clustering such as limited time and memory and one pass clustering. Furthermore, discovering clusters with arbitrary shapes is very important in data stream applications. Data streams are infinite and evolving over time, and we do not have any knowledge about the number of clusters. In a data stream environment due to various factors, some noise appears occasionally. Density-based method is a remarkable class in clustering data streams, which has the ability to discover arbitrary shape clusters and to detect noise. Furthermore, it does not need the number of clusters in advance. Due to data stream characteristics, the traditional density-based clustering is not applicable. Recently, a lot of density-based clustering algorithms are extended for data streams. The main idea in these algorithms is using density-based methods in the clustering process and at the same time overcoming the constraints, which are put out by data stream’s nature. The purpose of this paper is to shed light on some algorithms in the literature on density-based clustering over data streams. We not only summarize the main density-based clustering algorithms on data streams, discuss their uniqueness and limitations, but also explain how they address the challenges in clustering data streams. Moreover, we investigate the evaluation metrics used in validating cluster quality and measuring algorithms’ performance. It is hoped that this survey will serve as a steppingstone for researchers studying data streams clustering, particularly density-based algorithms.

Clustering of data with high dimension and variable densities poses a remarkable challenge to the traditional density-based clustering methods. Recently, entropy, a numerical measure of the uncertainty of information, can be used to measure the border degree of samples in data space and also select significant features in feature set. It was used in our new framework based on the sparsity-density entropy (SDE) to cluster the data with high dimension and variable densities. First, SDE conducts high-quality sampling for multidimensional data and selects the representative features using sparsity score entropy (SSE). Second, the clustering results and noises are obtained adopting a new density-variable clustering method called density entropy (DE). DE automatically determines the border set based on the global minimum of border degrees and then adaptively performs cluster analysis for each local cluster based on the local minimum of border degrees. The effectiveness and efficiency of the proposed SDE framework are validated on synthetic and real data sets in comparison with several clustering algorithms. The results showed that the proposed SDE framework concurrently detected the noises and processed the data with high dimension and various densities.

Trip end identification based on spatial-temporal clustering algorithm using smartphone positioning data

Hierarchical clustering, a traditional clustering method, has been getting attention again. Among several reasons, a credit goes to a recent paper by Dasgupta in 2016 that proposed a cost function that quantitatively evaluates hierarchical clustering trees. An important question is how to combine this recent advance with existing successful clustering methods. In this paper, we propose a hierarchical clustering method to minimize the cost function of clustering tree by incorporating existing clustering techniques. First, we developed an ensemble tree-search method that finds an integrated tree with reduced cost by integrating multiple existing hierarchical clustering methods. Second, to operate on large and arbitrary shape data, we designed an efficient hierarchical clustering framework, called integrating divisive and ensemble-agglomerate (IDEA) by combining it with advanced clustering techniques such as nearest neighbor graph construction, divisive-agglomerate hybridization, and dynamic cut tree. The IDEA clustering method showed better performance in minimizing Dasgupta's cost and improving accuracy (adjusted rand index) over existing cost-minimization-based, and density-based hierarchical clustering methods in experiments using arbitrary shape datasets and complex biology-domain datasets.

Density-based clustering is a strong family of clustering methods. The strength of this family is its ability to classify data of arbitrary shapes and to omit the noise. Among them density-based clustering (DENCLUE), which is one of the well-known powerful density-based clustering methods. DENCLUE is based on the concept of the hill climbing algorithm. In order to find the clusters, DENCLUE has to reach a set of points called density attractors. Despite the advantages of DENCLUE, it remains sensitive to the growth of the size of data and of the dimensionality, in the fact that the density attractors are calculated of each point in the input data. In this paper, in the aim to overcome the DENCLUE shortcoming, we propose an efficient approach. This approach replaces the concept of the density attractor by a new concept which is 'the hyper-cube representative'. The experimental results, provided from several datasets, prove that our approach finds a trade-off between the performance of clustering and the fast response time. In this way, the proposed clustering methods work efficiently for large of multidimensional data.

Historical information on credit card transactions can be used to generate a fraud score which can then be used to reduce credit card fraud. The report describes a fraud-nonfraud classification methodology using a radial basis function network (RBFN) with a density based clustering approach. The input data is transformed into the cardinal component space and clustering as well as RBFN modeling is done using a few cardinal components. The methodology has been tested on a fraud detection problem and the preliminary results obtained are satisfactory.

To prevent the problem of safety accidents caused by the intrusion of obstacles into railway clearance, this paper proposes an obstacle detection method based on Light Detection and Ranging (LiDAR) to obtain and process rich three-dimensional (3D) information and depth information of the railway scene. The method first preprocesses the point cloud of the railway scenario collected by LiDAR to divide a basic area containing the rails. Then, the method divides the roadbed plane and fits the rails with the random sample consensus (RANSAC) algorithm, dividing the detection area according to the position of the rails. To address the issue of over or under-segmentation in the traditional Euclidean clustering method, which is due to sparser point clouds the farther the object is from the LiDAR, this paper improves the conventional Euclidean clustering. It introduces an adaptive distance threshold to categorize obstacles. Finally, compared with traditional Euclidean clustering, K-means clustering, and density-based spatial clustering of applications with noise (DBSCAN) clustering, the improved Euclidean cluster has achieved better results in terms of computing time and segmentation accuracy. Experimental results show the ability of the method to detect railway obstacles successfully.

Recommendation systems are involved in many industries, for example (e-health, transportation, e-commerce, and agriculture), where Recommendation systems aim to benefit both market and user levels. They help consumers make the right decision based on their preferences without being exposed to data overload. Nowadays, there is a wide range of recommenders based on different filtering approaches, such as Collaborative-based, Content-based, hybrid-based, demographic-based filtering approaches. In this paper, we present clustering-based recommendation systems. We also experiment and show the results for a collaborative-based movie recommender using different clustering techniques such as Kmeans, BIRCH Balanced Iterative Reducing and Clustering using Hierarchies) and DBSCAN (Density-based Spatial Clustering of Applications with Noise). We intended to choose different clustering approaches such as partitional, hierarchical, and density-based clustering approaches. We incorporated Item-based Collaborative filtering, then applied multiple clustering techniques on the dataset based on the users' ratings. We checked the performance using accuracy measures such as MAE (Mean Absolute Error), RMSE (Root mean square error), and the computed time. These measures were calculated for analysis and comparison purposes.

Identification and assessment of sites for solar farms development using GIS and density based clustering technique- A case of Pakistan

Future self-driving cars and current ones with advanced driver assistance systems are expected to interact with other traffic participants, which often are multiple other vehicles. Object vehicle tracking forms a key part of resolving this interaction. Furthermore, descriptions of the vehicle group behaviors, like group formations or splits, can enhance the utility of the tracking information for further motion planning and control decisions. In this paper, we propose a probabilistic method to estimate the formation and evolution, including splitting, re-grouping, and so on, of object vehicle groups and the membership conditions for individual object vehicles forming the groups. A Bayesian estimation approach is used to first estimate the states of the individual vehicles in the presence of uncertainties due to sensor imperfections and other disturbances acting on the individual object vehicles. The closeness of the individual vehicles in both their positions and velocity is then evaluated by a probabilistic collision condition. Based on this, a density-based clustering approach is applied to identify the vehicle groups as well as the identity of the individual vehicles in each group. An estimation of the state of the group as well as of the group boundary is also given. Finally, detailed numerical experiments are included, including one on real-time traffic intersection data, to illustrate the workings and the performance of the proposed approach. The potential application of the approach in motion planning of autonomous vehicles is also highlighted.

Collaborative filtering recommenders leverage past user-item ratings in order to predict ratings for new items. One of the most critical steps in such methods corresponds to the formation of the neighbourhood that contains the most similar users or items, so that the ratings associated with them can be employed for predicting new ratings. This work proposes to perform the combination of content-based and ratings-based evidence during the neighbourhood formation step and thus identify the most similar neighbours in a hybrid manner. To this end, DBSCAN, a density-based clustering approach, is applied for identifying the most similar users or items by considering the ratings-based and the content-based similarities, both individually and in combination. The resulting hybrid cluster-based CF recommendation scheme is then evaluated on the latest small MovieLens100k dataset and the experimental results indicate the potential of the proposed approach.

The unveiling of communities within a network or graph, and the hierarchization of its members that results is of utmost importance in areas ranging from social to biochemical networks, from electronic circuits to cybersecurity. We present a statistical mechanics approach that uses a normalized Gaussian function which captures the impact of a node within its neighborhood and leads to a density-ranking of nodes by considering the distance between nodes as punishment. A hill-climbing procedure is applied to determine the density attractors and identify the unique parent (leader) of each member as well as the group leader. This organization of the nodes results in a tree-like network with multiple clusters, the community tree. The method is tested using synthetic networks generated by the LFR benchmarking algorithm for network sizes between 500 and 30,000 nodes and mixing parameter between 0.1 and 0.9. Our results show a reasonable agreement with the LFR results for low to medium values of the mixing parameter and indicate a very mild dependence on the size of the network.

Graph clustering aiming to partition nodes into several disjoint subsets is a fundamental task for graph-structured learning. Traditional graph clustering methods only consider the adjacency information. In recent years, inspired by the homophily assumption that the adjacent nodes tend to have similar features and labels, most existing graph clustering approaches leverage node attribute information to improve graph clustering performance. These works have mainly focused on node embedding learning via the various combinations of auto-encoder and graph neural networks. As for clustering learning, they introduce a self-optimizing strategy that assumes that all clusters are homogeneous. However, this assumption usually does not hold since the size and variance of different clusters can be quite different, and self-optimizing strategy is incompetent in dealing with this heterogeneous clusters. In this work, we propose a novel method named Adaptive Harmony Learning and Optimization (AHLO) for attributed graph clustering, which models the node embeddings with the mixture of von Mises-Fisher distributions on the unit hypersphere and develops an alternating learning strategy. Specifically, we take the node embeddings as the supervisory signals for the update of the mixture parameters, and the mixture distribution as the supervisory signals for the update of the node embeddings. To prevent small clusters from annexing by large clusters, we develop the regularized harmony loss to enhance the prediction on small clusters. In the mixture parameter optimization stage, we utilize EM algorithm and heuristically design a center update scheme with consideration of the posterior probability confidence and the impact of other centers. Hence, AHLO can simultaneously improve the intra-cluster compactness and inter-cluster separability. Extensive experiments on four benchmark attributed graph datasets have demonstrated the effectiveness of our proposed AHLO.

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.

In this letter, we propose a novel method to reconstruct accurate building structures from high-resolution multiview aerial imagery, using layered contour fitting (LCF) with a density-based clustering algorithm. Initially, the complicated 3-D scene is reconstructed by a probabilistic volumetric modeling algorithm. Subsequently, the reconstructed 3-D scene model is projected into layer images based on the height information. At last, we combine an extended layered density-based clustering approach with a generative LCF approach to remove noise and extract accurate building contours in every layer image at the same time. The final accurate 3-D building model is generated from these contours in layer images with a smoothing operation. Experiments on the aerial image sets demonstrate effectiveness and precision of our method.