Compatible Cluster Research Articles

Deep embedding for mental health content on social media using vector space model with feature clusters

AbstractOver the years, various document‐clustering techniques were developed to group the textual data. The performance of document clustering systems heavily relies on the optimal use of text representations. Vector space model is an extensively used technique by existing clustering algorithms to present the text in a structured form. However, such representations suffer from a lack of semantic associations, high dimensionality, and sparsity. In order to enrich the document representation by retaining the semantic and morphological associations, this paper introduced a word cluster‐based modified term frequency‐inverse document frequency (WC_MTI) model, in which semantically associated word embeddings from the word2vec are supplemented with morphological information using kernel principal component analysis. In addition, to address high dimensionality and sparsity issues and improve the clustering, we use a self‐training technique that learns discriminative features using the WC_MTI model and autoencoder (AE) and then updates the encoder network weights using assignments from a clustering algorithm as supervision. The proposed model organizes documents into topically compatible clusters by maintaining the semantic and morphological similarity between terms using the skip‐gram with negative sampling (SGNS) and low‐dimensional vector representations. We evaluate the proposed approach against the existing text representation methods. Experimental findings suggest that the proposed approach enhanced the average accuracy by 89.62%.

A multi-sector model of relatedness, growth and industry clustering

AbstractThis article builds an understanding of regional innovation specialisation by developing a multi-sector model with endogenous growth through quality improving innovations and spillovers from related technologies. The model provides an approach to incorporate the relatedness literature within the mainstream theoretical frameworks of endogenous growth and economic geography. Each firm’s technology sector and the location of other firms play a role in each firm’s ability to improve its own technology. As a result, firms prefer to co-locate in technologically compatible clusters. Without relying on scale assumptions, the model for the first time coherently links related variety knowledge spillovers to mainstream urban economic frameworks and demonstrates that clustering is possible in both core and peripheral areas.

Minimization of Energy Consumption in WSN using Hybrid WECRA Approach

In wireless sensor network, the energy efficiency protocols are utilized to improve the energy conservation of the system and enhance the lifetime. The transmission is done by the deployment of nodes instead of any physical medium for communication in WSN. The deployed nodes transmit the data to sink or to the destination node. The nodes are gathered for better communication and this is known as clustering. In clustering, the nearby nodes are situated in compatible cluster and send the information to the cluster head which aggregates the information from various nodes and send it further to the sink. Different clustering algorithms and techniques have been developed for the enhancement of WSN. In traditional approach, the central node in the network preferred as cluster head and this node constrained for network transmission. This single node consumes more energy rather than other nodes leads to dead node. On the other hand, it follows the straight line route therefore the route is overloaded which consumes more energy with the reduction in lifetime of the network. We proposed an energy efficient genetic algorithm based approach with the concept of Virtual Grid based Dynamic Routes Adjustment (VGDRA) which enhances the overall performance of wireless sensor networks. The proposed approach has better energy efficiency as compare to LEACH because it is dynamic approach not static, balanced the load and optimization creates more chance of better result in less number of loops that are not possible in other techniques. The simulation result of the proposed approach is carried out in MATLAB.

A Novel Service-orientated Incentive Compatible Reputation Model for MANETs

Mobile Ad Hoc networks have become increasingly popular in the recent years. Multi-devices cooperating and sharing services is critical to realizing various mobile business applications. However, because mobile nodes are usually constrained by limited resources such as energy, bandwidth, computational power, storage and so on, some nodes may be a unwilling to share the services. Under these conditions we present in this paper a novel trust and reputation model, called NSOICRep, based on an incentive compatible clustering global reputation model aiming at providing trust and reputation in MANETs. Simulations show that NSOICRep can resist the selfish behavior of nodes, and motivate nodes positively share trustworthy services effectively.

Sparse Manifold Clustering and Embedding to discriminate gene expression profiles of glioblastoma and meningioma tumors

Sparse Manifold Clustering and Embedding (SMCE) algorithm has been recently proposed for simultaneous clustering and dimensionality reduction of data on nonlinear manifolds using sparse representation techniques. In this work, SMCE algorithm is applied to the differential discrimination of Glioblastoma and Meningioma Tumors by means of their Gene Expression Profiles. Our purpose was to evaluate the robustness of this nonlinear manifold to classify gene expression profiles, characterized by the high-dimensionality of their representations and the low discrimination power of most of the genes. For this objective, we used SMCE to reduce the dimensionality of a preprocessed dataset of 35 single-labeling cDNA microarrays with 11500 original clones. Afterwards, supervised and unsupervised methodologies were applied to obtain the classification model: the former was based on linear discriminant analysis, the later on clustering using the SMCE embedding data. The results obtained using both approaches showed that all (100%) the samples could be correctly classified and the results of all repetitions but one formed a compatible cluster of predictive labels. Finally, the embedding dimensionality of the dataset extracted by SMCE revealed large discrimination margins between both classes.

Compatible Clustering Algorithm with Convex Space Partitioning

In this paper, we discuss the problem of compatible clustering, we propose a new compatible clustering algorithm based on pNCompClu[9]. The new algorithm adopts space partitioning technique to replace the point neighborhood mechanism of pNCompClu. Experiments show that the proposed algorithm can get some consistent clustering results, and theory analysis also demonstrates that the proposed algorithm has higher clustering precision than pNCompClu has.

Remote Plasma Processing of Sapphire Substrates for Deposition of TiN and TiO<SUB>2</SUB>

The paper uses remote plasma assisted deposition, oxidation and nitridation processes for depositing thin films of metallic TiN on crystalline sapphire (0001) substrates. These films on sapphire substrates are being studied as window materials for high power radio frequency (RF) power tubes. A sequence of four process steps has been performed in a reactor chamber that isolates the deposition and surface-processing chamber from the plasma generation region. The chamber is part of an ultra-high-vacuum (UHV) compatible multi-chamber cluster in which the sequence of four process steps can be interrupted after each step, and surface chemistry changes can be identified by in-line Auger electron spectroscopy (AES). The four process steps, performed after an ex-situ chemical clean and blow-dry in nitrogen gas, are (i) a remote plasma-assisted oxidation (RPAO) in which surface contaminants including adventitious carbon are removed; (ii) a remote plasma-assisted nitridation (RPAN) process which forms a superficial layer of generic AION used to increase surface adhesion of the TiN films; (iii) a remote plasma-enhanced chemical-vapor deposition (RPECVD) process for deposition of 2 to 5 nm thick TiN films, and finally (iv) a second RPAN step that increases the ratio of Ti-N bonding in the TiN films with respect to adventitious O-atom incorporation from the Ti precursor, Ti tetra-butoxide.

Majority-rule (+) consensus trees

The construction of a consensus tree to summarize the information of a given set of phylogenetic trees is now routinely a part of many studies in systematic biology. One popular method is the majority-rule consensus tree. In this paper we introduce and characterize a new consensus method that refines the majority-rule tree by adding certain compatible clusters satisfying a simple criterion.

Construction and Application of a UHV Compatible Cluster Deposition System

The design and operation of a new UHV-compatible atomic cluster deposition system is described. The design is optimised for high cluster fluxes and for the production of cluster-assembled nano-devices. One key feature of the system is a high degree of flexibility, including interchangeable sputtering and inert gas aggregation sources, and two kinds of mass spectrometer, which allow both characterisation of the cluster size distribution and deposition of mass-selected clusters. Another key feature is that clusters are deposited onto electrically contacted lithographically defined devices mounted on an UHV-compatible cryostat cold finger, allowing deposition at room temperature as well as cryogenic and elevated temperatures. In-situ electrical characterisation of cluster-assembled devices can then be performed in the temperature range 1.2--75 K.

Retail concentration: a comparison of spatial convenience in shopping strips and shopping centres

The emergence of the Internet and a more discerning consumer has created the need for traditional retail centres to provide a more convenient shopping environment. A retail centre offers convenience when it minimises the spatial, temporal and effort costs of shopping. Existing strategies for spatial convenience include limiting the size of a retail centre by controlling the entry of non-retail firms, creating a compact physical design, and creating compatible clusters of shops. The authors’ propose an alternative method; the degree of retail concentration. The purpose of this study is to determine the extent to which shopping centres (also referred to as shopping malls) and shopping strips (also referred to as the high street, downtown or city centre) provide retail concentration. From the measures taken, this study provides insight into the degree of retail concentration offered by a sample of nine shopping centres (also referred to as a shopping mall) and nine shopping strips (also referred to as the high street, downtown or city centre). The findings yielded three important insights. Firstly, across the three measures of retail concentration, the shopping centre was found to offer consumers’ greater spatial convenience. Secondly, the findings add support to the notion that the demise of the shopping strip could be linked to its inability to satisfy the needs of a convenience-oriented society. And thirdly, while the shopping strip may be at a competitive disadvantage in terms of spatial convenience, market mechanisms such as Bid Rent Theory provided a better-than-expected spatial juxta-positioning of its businesses.

Centroid neural network for unsupervised competitive learning

An unsupervised competitive learning algorithm based on the classical -means clustering algorithm is proposed. The proposed learning algorithm called the centroid neural network (CNN) estimates centroids of the related cluster groups in training date. This paper also explains algorithmic relationships among the CNN and some of the conventional unsupervised competitive learning algorithms including Kohonen's self-organizing map (SOM) and Kosko's differential competitive learning (DCL) algorithm. The CNN algorithm requires neither a predetermined schedule for learning coefficient nor a total number of iterations for clustering. The simulation results on clustering problems and image compression problems show that CNN converges much faster than conventional algorithms with compatible clustering quality while other algorithms may give unstable results depending on the initial values of the learning coefficient and the total number of iterations.

A robust algorithm for automatic extraction of an unknown number of clusters from noisy data

A clustering algorithm that combines the advantages of fuzzy clustering and robust statistical estimators is presented. The algorithm generates two sets of weights (memberships) for each feature vector as a by-product. These two sets of weights are used to partition the data set and to obtain robust estimates of the prototype parameters, respectively. An extension of the algorithm to deal with an unknown number of clusters is also proposed. The extension is based on overspecifying the number of clusters, and merging compatible clusters.

Fitting an unknown number of lines and planes to image data through compatible cluster merging

A compatible cluster merging algorithm is presented that is specially designed to find the optimum number of linear, planar and hyperplanar clusters (i.e. clusters that lie in a subspace of the original space), when an upper bound on the number of clusters present is known. This algorithm is shown to be superior to more traditional validity-measure-based approaches. The effectiveness and advantages of the proposed technique in 2D and 3D applications is demonstrated with both synthetic and real data. The proposed applications include character recognition, obtaining straight-line descriptions of intensity edge images and obtaining planar approximations of 3D (range) data.

The fuzzy c spherical shells algorithm: A new approach

The fuzzy c spherical shells (FCSS) algorithm is specially designed to search for clusters that can be described by circular arcs or, generally, by shells of hyperspheres. A new approach to the FCSS algorithm is presented. This algorithm is computationally and implementationally simpler than other clustering algorithms that have been suggested for this purpose. An unsupervised algorithm which automatically finds the optimum number of clusters is not known. It uses a cluster validity measure to identify good clusters, merges all compatible clusters, and eliminates spurious clusters to achieve the final results. Experimental results on several data sets are presented.