Proposed Clustering Model Research Articles

Self-Tuning Possibilistic c-Means Clustering Models

The relaxation of the probabilistic constraint of the fuzzy c-means clustering model was proposed to provide robust algorithms that are insensitive to strong noise and outlier data. These goals were achieved by the possibilistic c-means (PCM) algorithm, but these advantages came together with a sensitivity to cluster prototype initialization. According to the original recommendations, the probabilistic fuzzy c-means (FCM) algorithm should be applied to establish the cluster initialization and possibilistic penalty terms for PCM. However, when FCM fails to provide valid cluster prototypes due to the presence of noise, PCM has no chance to recover and produce a fine partition. This paper proposes a two-stage c-means clustering algorithm to tackle with most problems enumerated above. In the first stage called initialization, FCM with two modifications is performed: (1) extra cluster added for noisy data; (2) extra variable and constraint added to handle clusters of various diameters. In the second stage, a modified PCM algorithm is carried out, which also contains the cluster width tuning mechanism based on which it adaptively updates the possibilistic penalty terms. The proposed algorithm has less parameters than PCM when the number of clusters is [Formula: see text]. Numerical evaluation involving synthetic and standard test data sets proved the advantages of the proposed clustering model.

Mixture of Joint Nonhomogeneous Markov Chains to Cluster and Model Water Consumption Behavior Sequences

The emergence of smart meters has fostered the collection of massive data that support a better understanding of consumer behaviors and better management of water resources and networks. The main focus of this article is to analyze consumption behavior over time; thus, we first identify the main weekly consumption patterns. This approach allows each meter to be represented by a categorical series, where each category corresponds to a weekly consumption behavior. By considering the resulting consumption behavior sequences, we propose a new methodology based on a mixture of nonhomogeneous Markov models to cluster these categorical time series. Using this method, the meters are described by the Markovian dynamics of their cluster. The latent variable that controls cluster membership is estimated alongside the parameters of the Markov model using a novel classification expectation maximization algorithm. A specific entropy measure is formulated to evaluate the quality of the estimated partition by considering the joint Markovian dynamics. The proposed clustering model can also be used to predict future consumption behaviors within each cluster. Numerical experiments using real water consumption data provided by a water utility in France and gathered over 19 months are conducted to evaluate the performance of the proposed approach in terms of both clustering and prediction. The results demonstrate the effectiveness of the proposed method.

A Cluster-Based Operation Model of Aggregated Battery Swapping Stations

Battery swapping stations (BSSs) have a potential of becoming a part of the electric vehicle charging infrastructure mix. A number of aggregated BSSs have a significant capacity and can act as strategic players in electricity market. This paper formulates a discrete cluster model for optimal operation of aggregated BSSs that take part in the day-ahead energy and reserve capacity markets and have sufficient combined capacity to affect market prices. In the presented bilevel structure the upper-level problem optimally defines charging and discharging schedules, as well as reserve capacity, of the aggregated BSSs, while the lower-level problem clears the market and provides locational marginal prices for each BSS, as well as the reserve capacity prices. This problem is converted into a single-level equivalent using the Karush-Kuhn-Tucker optimality conditions. The presented case study reveals that BSSs can collect significant revenues when providing reserve capacity. The main benefit of the proposed cluster model is computational tractability and solution accuracy, which depends on the resolution of the clusters.

A Clustering Based Classification Approach Based on Modified Cuckoo Search Algorithm

Cuckoo Search Algorithm (CSA) is one of the new swarm intelligence based optimization algorithms, which has shown an effective performance on many optimization problems. However, the effectiveness of CSA significantly depends on the exploration and exploitation potential and it may also possible to increase its efficiency when solving complex optimization problems. In this study, some mechanisms have been employed on CSA to increase its efficiency such as use of global best and individual best solutions to guide the other solutions, self-adaption techniques for parameters and so on. The modified CSA (i.e., MCSA) is successfully employed in clustering based classification domain. The experimental results and execution time prove its effectiveness over existing modified CSAs and other employed swarm intelligence algorithms. The proposed clustering model is also employed in color histopathological image segmentation domain and provides effective result.

EELCM: An Energy Efficient Load-Based Clustering Method for Wireless Mobile Sensor Networks

Clustering techniques are employed to optimize energy consumption corresponding to different node activities and thus to conserve the limited energy of the sensor nodes. Different locations and different loads cause unbalanced energy consumption in the network nodes, consequently, the network lifetime is shortened. In this paper, a clustering method to improve energy conserving and to balance the energy consumption of cluster head is proposed. The area and the dimensions of the clusters are calculated by the algorithm in such a way that the rates of energy consumption of all the cluster heads be equalized. Also, all the intra-cluster communications are performed in single hop manner. The proposed approach presents a solution for the challenge of short lifetime of heavy burdened cluster heads around the base station. A straightforward routing scheme is proposed corresponding to the proposed clustering model to deliver the data sensed by cluster members to the base station. The performance evaluations indicate that the proposed method not only prolongs more effectively the lifetime of both cluster heads and cluster members than the other works, but also, outperforms them in balancing the energy level of the cluster heads. Consequently, the duration of activity of the cluster members in different clusters is also balanced and prolonged. Moreover, the proposed method improves the network throughput, comparing with the other works.

Static malware clustering using enhanced deep embedding method

SummaryMalware refers to any software, programs, or files that are intentionally utilised to compromise the system and cause unexpected losses to end‐users such as economical losses or privacy breaches. The rapid growth of malware makes it impossible to keep up with its progress merely via human interventions or manual analysis. One of the challenges for the human‐oriented approaches is they will cause backlog and inability to keep up with the development traces of the malware. Hence, an efficient method is needed urgently to analyse effectively and identify accurately the malware in their domain. Malware clustering has been extensively studied in the machine learning area with regards to distance functions, grouping algorithm and cluster validation. A large number of research studies have been done via behavioral analysis for clustering to achieve high performance of malware detections. However, there is a trade‐off for better detection performance between behaviorial approaches and high computational forces. Up to date, little work focuses on the deep learning representations for malware clustering. Therefore, in this paper, we propose an enhanced deep embedded clustering method to facilitate an effective and efficient malware clustering process. The new method takes advantage of linear dimensionality reduction and a customised deep neural network to learn malware representations in an orthogonal space and performs cluster assignments. Our experimental results demonstrate that the proposed clustering model outperforms the traditional K‐means method with regards to the enhanced features using various auto‐encoder, pre‐trained weight and principle component analysis (PCA).

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

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

Configuration-Based Fingerprinting of Mobile Device Using Incremental Clustering

Device fingerprinting has lately received great attention due to its effectiveness in fraud detection, secure authentication, and user tracking. Whereas fingerprinting performs well on labeled device data using classification methods, there are several researches concentrated on unlabeled mobile device data and existing methods often lack in precision and efficiency. To overcome this challenge, we focus on the use of the mobile device’s configuration-related characteristics as a mean to build a device fingerprint, which allows to distinctively and reliably characterize each device. In addition, an incremental clustering approach is proposed to classify unlabeled device data into clusters on the basis of their similarity. Moreover, we customize individual distance threshold for each user according to their device configurations’ modifying frequency, in order to construct a precise authentication mechanism between users and devices. The proposed clustering model and device fingerprinting system are evaluated on 8220 fingerprints from 815 different devices. The experimental results demonstrate the effectiveness and efficiency of our algorithms.

Speeding up the large-scale consensus fuzzy clustering for handling Big Data

Massive data can create a real competitive advantage for the companies; it is used to better respond to customers, to follow the behavior of consumers, to anticipate the evolutions, etc. However, it has its own deficiencies. This data volume not only requires big storage spaces but also makes analysis, processing and retrieval operations very difficult and hugely time-consuming. One way to overcome these problems is to cluster this data into a compact format that is still an informative version of the entire data. A lot of clustering algorithms have been proposed. However, their scaling is poor in terms of computation time whenever the size of the data gets larger. In this paper, we make full use of consensus clustering to handle Big Data clustering. We use sampling combined with a split-and-merge strategy to fragment data into small subsets, then basic partitions are locally generated from them using RHadoop's parallel processing MapReduce model and later a consensus tendency is followed to obtain the final result. A scalability analysis is conducted to demonstrate the performance of the proposed clustering models by increasing both the number of computing nodes used and the sample size while satisfying the volume and the velocity dimensions.

Hybrid self-optimized clustering model based on citation links and textual features to detect research topics.

The challenge of detecting research topics in a specific research field has attracted attention from researchers in the bibliometrics community. In this study, to solve two problems of clustering papers, i.e., the influence of different distributions of citation links and involved textual features on similarity computation, the authors propose a hybrid self-optimized clustering model to detect research topics by extending the hybrid clustering model to identify “core documents”. First, the Amsler network, consisting of bibliographic coupling and co-citation links, is created to calculate the citation-based similarity based on the cosine angle of papers. Second, the cosine similarity is also used to compute the text-based similarity, which consists of the textual statistical and topological features. Then, the cosine angle of the linear combination of citation- and text-based similarity is considered as the hybrid similarity. Finally, the Louvain method is applied to cluster papers, and the terms based on term frequency are used to label clusters. To test the performance of the proposed model, a dataset related to the data envelopment analysis field is used for comparison and analysis of clustering results. Based on the benchmark built, different clustering methods with different citation links or textual features are compared according to evaluation measures. The results show that the proposed model can obtain reasonable and effective clustering results, and the research topics of data envelopment analysis field are also analyzed based on the proposed model. As different features are considered in the proposed model compared with previous hybrid clustering models, the proposed clustering model can provide inspiration for further studies on topic identification by other researchers.

Weighted Graph Based Clustering and Local Mobility Management for Dense Small Cell Network with X2 Interface

Along with the surge in mobile data, dense small cell network has become an effective method to improve system capacity and spectrum efficiency. However, because more small cells are deployed, the interference among dense small cells exacerbates. It also makes frequent handover for mobile users (UEs), which brings a great deal of signaling overhead to the core network. In order to solve the problems of interference and frequent handover, a novel clustering scheme for dense small cell network is proposed in this paper. The scheme is based on the weighted graph. First, we present a dense small cell clustering model based on X2 interface to minimize core network signaling overhead. To improve the usability of the model, we model the system as an undirected weighted graph. Then we propose the maximum benefit merging algorithm to reduce the complexity. This method enables adjacent small cells to cooperate and form virtual cellular cluster according to handover statistics information. Then we select cluster head (CH) according to certain rule in each cluster. Cluster head acts as the mobility anchor, managing the handovers between cluster members. This can reduce core network signaling overhead and the interference among small cells effectively. Compared with the 3GPP handover algorithm, the proposed clustering model in this paper can reduce the signaling overhead more than 70%. The simulation results show that the proposed clustering model can effectively cluster the dense small cell.

COCUS: Concept Based Document Clustering by Corpus Utility Scale

Objective: With the rising quantum of documents in corpuses, it is very important that data management and data assurance is with high interoperability towards retrieving the critical documents from vast range of services. By focusing on the semantic features, which could improve the level of accuracy in document tracing and retrieval, the issues and limitations in the present models could be addressed in an effective manner. Methods/Statistical Analysis: In this study, focus is on depicting the robustness of semantic features based clustering techniques and its efficacy, compared to the other kind of clustering techniques. This paper proposed a concept based document clustering by corpus utility scale (COCUS) proposed. The utility scale proposed in COCUS is derived with support of topic related selected document set as knowledge base that enables to cluster the documents by their concept relevancy. The proposed clustering model is assessed through the state of the art metrics called cluster purity, inverse of purity and cluster level harmonic mean. Experiments were carried out on datasets that comprise the containing specific kind of literature gathered from varied open access journals from publishers. The total 1509 number of documents was collected and among them 497 documents was used as knowledgebase and rest 1012 documents were used for clustering process. Findings: The experimental study evincing that the proposed model is scalable and robust. The purity and harmonic mean of the resultant clusters confirming that the COCUS clusters the documents by their concept relevancy with 94% accuracy (Average of the topic level harmonic mean of the clusters was found as 0.94). Application/ Improvements: The computational complexity of the COCUS is evinced as linear, where the majority of benchmarking models are found to be np-hard.

Autoregressive metric-based trimmed fuzzy clustering with an application to PM10 time series

Air quality measurement relies on the effectiveness of a network of monitoring stations. Monitoring stations collect information about the evolution of air pollutants concentration. If more stations supplies the same information, then some of them could be deemed as redundant. Then, a clustering model for time series is useful to identify stations with similar features. Time series of pollutant concentration can be classified using the autoregressive metric in the framework of standard clustering techniques. A serious drawback is related to the lack of robustness of standard procedures. In this paper, using a partitioning around medoids approach combined with a trimming-based rule, a fuzzy model for cluster time series is proposed. The model provides a robust alternative to standard procedures. Two simulation studies are carried out to evaluate the clustering performance of the proposed clustering model. Finally, an empirical application to real time series of PM10 concentration in the Lazio region is presented and discussed showing the practical usefulness of the proposed approach.

Alignment of CH3F in para-H2 crystal studied by IR quantum cascade laser polarization spectroscopy.

In order to investigate the alignment of CH3F in para-H2 crystals, high resolution polarization spectroscopy of the ν3 vibrational band is studied using a quantum cascade laser at 1040 cm(-1). It is found that the main and satellite series of peaks in the ν3 vibrational band of CH3F have the same polarization dependence. This result supports the previously proposed cluster model with ortho-H2 in first and second nearest neighbor sites. The observed polarization dependence function is well described by a simple six-axis void model in which CH3F is not aligned along the c-axis of the crystal but tilted to 64.9(3)° from it.

User-Centric Cross-Tier Base Station Clustering and Cooperation in Heterogeneous Networks: Rate Improvement and Energy Saving

Heterogeneous cellular networks (HetNets) are to be deployed for future wireless communication to meet the ever-increasing mobile traffic demand. However, the dense and random deployment of small cells and their uncoordinated operation raise important concerns about various costs issues, among which notably is energy efficiency. Base station (BS) cooperation is set to play a key role in managing interference in HetNets. In this paper, we consider BS cooperation in the downlink HetNets where BSs from different tiers within the respective cooperative clusters jointly transmit the same data to a typical user, and in particular focus on the optimization of the energy efficiency performance. First, based on a proposed clustering model, we derive the spectral efficiency using tools from stochastic geometry. Furthermore, we formulate a power minimization problem with a minimum spectral efficiency constraint and derive the optimal received signal strength (RSS) thresholds under certain approximation. Building upon these results, we could address the problem of how to design appropriate RSS thresholds, taking into account the tradeoff between spectral efficiency and energy efficiency. Simulations show that the proposed clustering model is more energy-saving than the geometric clustering model, and deploying a multitier HetNet is significantly more energy-saving compared to a macro-only network.

Clustering regression based on interval-valued fuzzy outputs and interval-valued fuzzy parameters

In this paper, a new approach is presented to the problem of the clustering regression models with imprecise quantities. In this approach, the response variable and the parameters of model are assumed to be the interval-valued fuzzy numbers. We introduce two indices to investigate the goodness-of-fit of such models based on the similarity measure and the squared errors. In addition to, the predictive ability of the proposed clustering models is evaluated by using the cross-validation method. Finally, the application of the proposed approach in modeling some soil characteristics is studied.

GARCH-based robust clustering of time series

In this paper we propose different robust fuzzy clustering models for classifying heteroskedastic (volatility) time series, following the so-called model-based approach to time series clustering and using a partitioning around medoids procedure. The proposed models are based on a GARCH parametric modeling of the time series, i.e. the unconditional volatility and the time-varying volatility GARCH representation of the time series. We first suggest a timid robustification of the fuzzy clustering. Then, we propose three robust fuzzy clustering models belonging to the so-called metric, noise and trimmed approaches, respectively. Each model neutralizes the negative effects of the outliers in the clustering process in a different manner. In particular, the first robust model, based on the metric approach, achieves its robustness with respect to outliers by taking into account a “robust” distance measure; the second, based on the noise approach, achieves its robustness by introducing a noise cluster represented by a noise prototype; the third, based on the trimmed approach, achieves its robustness by trimming away a certain fraction of outlying time series. The usefulness and effectiveness of the proposed clustering models is illustrated by means of a simulation study and two applications in finance and economics.

An Energy Efficient Cluster Selection Optimization using Evolutionary Imperialist Competitive Algorithm

Wireless Communications is one of the fastest growing segments in the communications industry. Wireless Network is the network that facilitates communication among two or more devices connected through the standard network protocols, without network cabling. Due to the battery constrained the network performance will get reduced i.e. If the energy of the wireless sensor node (WSN’s) is drained, recharging of the sensor nodes in unattended environment is very difficult. As WSN nodes are usually battery-powered devices, the important and most critical aspects to face concern is how to reduce the energy consumption of WSN nodes, so that the network lifetime can be enhanced to an extent. Routing the Data in sensor nodes plays a vital role in transferring the data to the base station (BS). Different types of routing algorithm have been used such multihopping, grid based, hierarchical based and clustering based such LEACH, HEED etc... In this we have focused on incorporating clustering technique based on evolutionary technique namely ICA cluster optimization to improve the lifetime of the sensor nodes. We compare our proposed clustering model with LEACH protocol and analyze its efficiency.

Time series clustering by a robust autoregressive metric with application to air pollution

In this paper, following a fuzzy approach and adopting an autoregressive parameterization, we propose a robust clustering model for classifying time series. In particular, by adopting a fuzzy partitioning around medoids approach, the suggested clustering model is able to define the so-called medoid time series, which is a representative time series of each cluster, and the membership degrees of each time series to the different clusters. The robustness of the proposed clustering model is guaranteed by the adoption of a suitable robust metric for time series, i.e. the so-called exponential distance measure. In this way, the clustering model is able to tolerate the presence of outlier time series in the clustering process. In particular, it is capable of neutralizing and smoothing the disruptive effect of outlier time series, preserving the original clustering structure of the dataset, by assigning to outlier time series approximately the same membership degrees across clusters. To illustrate the usefulness and effectiveness of the suggested time series clustering model, a simulation study and an application to air pollution time series are carried out. Comparison with some existing clustering procedures suggested in the literature shows several advantages of the proposed model.

확률론적 클러스터링 기법을 이용한 한반도 태풍경로 및 종관기후학적 분석

최근 빈번하게 발생하는 태풍사상은 극심한 홍수 및 바람 재해를 유발 시키고 있다. 이러한 점에서 본 연구에서는 1951년부터 2012년까지 한반도에 내습한 총 197개의 태풍사상을 대상으로 태풍의 발생위치 및 태풍의 궤적을 기준으로 태풍을 범주화 할 수 있는 확률론적 클러스터링 기법을 개발하였다. 모의실험을 통하여 개발된 모형의 적합성을 확인할 수 있었으며, 태풍 경로에 적용이 가능한 방안으로 평가되었다. 1951년부터 2012년까지 한반도 내습한 197개의 태풍사상을 대상으로 확률론적 클러스터링 기법을 적용한 결과 한반도를 내습한 태풍사상은 총 7개의 클러스터로 분류되었으며, 대부분 위도 <TEX>$10^{\circ}{\sim}20^{\circ}N$</TEX>, 경도 <TEX>$120^{\circ}{\sim}150^{\circ}E$</TEX> 해수면에서 발생하여 한반도를 향하여 진행하는 것으로 나타났다. 클러스터 B의 경우 약 25.4%의 발생빈도를 가지며, 전선의 방향도 한반도를 직접 향하고 있어 상대적으로 한반도에 영향이 가장 큰 클러스터로 분석되었으며 한반도 전체에 걸쳐서 강한 양(positive)의 강우량 Anomaly를 갖는 것을 확인할 수 있었다. Lately, more frequent typhoons cause extensive flood and wind damage throughout the summer season. In this respect, this study aims to develop a probabilistic clustering model that uses both typhoon genesis location and trajectories. The proposed model was applied to the 197 typhoon events that made landfall in the Korean peninsula from 1951 to 2012. We evaluate the performance of the proposed clustering model through a simulation study based on synthetic typhoon trajectories. The seven distinguished clusters for typhoons affecting Korean peninsula were identified. It was found that most of typhoon genesis originated from a remote position (<TEX>$10^{\circ}{\sim}20^{\circ}N$</TEX>, <TEX>$120^{\circ}{\sim}150^{\circ}E$</TEX>) near the Equator. Cluster, type B can be regarded as a major track due to the fact that its frequency is approximately about 25.4% out of 197 events and its direct association with strong positive rainfall anomalies.