Incorporating Clustering Techniques into GAMLSS

A method for statistical analysis of multimodal and/or highly distorted data is presented. The new methodology combines different clustering methods with the GAMLSS (generalized additive models for location, scale, and shape) framework, and is therefore called c-GAMLSS, for “clustering GAMLSS. ” In this new extended structure, a latent variable (cluster) is created to explain the response-variable (target). Any and all parameters of the distribution for the response variable can also be modeled by functions of the new covariate added to other available resources (features). The method of selecting resources to be used is carried out in stages, a step-based method. A simulation study considering multiple scenarios is presented to compare the c-GAMLSS method with existing Gaussian mixture models. We show by means of four different data applications that in cases where other authentic explanatory variables are or are not available, the c-GAMLSS structure outperforms mixture models, some recently developed complex distributions, cluster-weighted models, and a mixture-of-experts model. Even though we use simple distributions in our examples, other more sophisticated distributions can be used to explain the response variable.

The immunoglobulin kappa variable cluster is differentially expressed in human epithelial ovarian cancer.

Epithelial ovarian cancer (EOC) is the most lethal gynecologic cancer (1). We performed discovery of genes associated with epithelial ovarian cancer and of the high-grade serous ovarian cancer (HGSC) subtype, using published and public microarray data (2, 3) to compare global gene expression profiles of normal ovary or fallopian tube with that of primary tumors from women diagnosed with epithelial ovarian cancer or HGSC. We identified genes encoded by the immunoglobulin kappa variable cluster, IGKV, as among the genes whose expression was most different in epithelial ovarian cancer as compared to the normal fallopian tube. IGKV1D-8 expression was significantly higher in high-grade serous ovarian tumors relative to normal fallopian tube. These data indicate that expression of IGKV genes is perturbed in epithelial ovarian cancers broadly and in ovarian cancers of the HGSC subtype. IGKV may be relevant to pathways underlying ovarian cancer initiation (transformation) or progression.

EUCOR: An Efficient Unequal Clustering and Optimal Routing in wireless sensor networks for energy conservation

Wireless sensor network is a wide network that works as a cutting edge model in industrial applications. The sensor application is mostly used for high security systems that provide safety support to the environment. The sensor system senses the physical phenomenon, processes the input signal and communicates with the base station through its neighbors. Energy is the most important criterion to support a live network for long hours. In the proposed system, the EUCOR (Efficient Unequal Clustering and Optimized Routing) protocol uses the objective function to identify the efficient cluster head with variable cluster size. The computation of the objective function deals with the ant colony approach for minimum energy consumption and the varying size of the cluster in each cycle is calculated based on the competition radius. The system prolongs the lifespan of the nodes by minimizing the utilization of energy in the transmission of packets in the networks when compared with the existing system.

Generation of time-domain-multiplexed two-dimensional cluster state

Entanglement is the key resource for measurement-based quantum computing. It is stored in quantum states known as cluster states, which are prepared offline and enable quantum computing by means of purely local measurements. Universal quantum computing requires cluster states that are both large and possess (at least) a two-dimensional topology. Continuous-variable cluster states—based on bosonic modes rather than qubits—have previously been generated on a scale exceeding one million modes, but only in one dimension. Here, we report generation of a large-scale two-dimensional continuous-variable cluster state. Its structure consists of a 5- by 1240-site square lattice that was tailored to our highly scalable time-multiplexed experimental platform. It is compatible with Bosonic error-correcting codes that, with higher squeezing, enable fault-tolerant quantum computation.

A Factorial Study of Bottlenecks in Nigerian Petroleum Refineries

Bottlenecks in the refineries lead to the disruption of refinery operations which result in production loss and time wastage. Nigerian refineries are four and they have not been able to work optimally as they have failed to produce up to their installed capacity. A lot of factors are contributing to this and are known as bottlenecks. This study was taken so as to identify those bottlenecks in the refineries with a view of making them known so that actions can be taken to tackle them and get Nigerian Refineries move from their pariah states to a welcome state. Kendall’s Coefficient of Concordance (K.C.C) and Principal Component Analysis (P.C.A) which are tools in factor analysis were employed. The K.C.C helped in ranking the identified variables according to their order of importance while the PCA helped to achieve parsimony through factor reduction. The results obtained revealed that the experts ranking of the thirty two scale items were in agreement at an alpha level of 0.01 and the computed coefficient of concordance was 0.51which is substantial. The thirty two scale items were able to be reduced into mere five clusters by PCA. A lone variable cluster which was labeled creatively ‘Government interference’ came up trump and account for most of the challenges being experienced in the Refineries. Other clusters labeled creatively were Eclectic issues, organizational management, Supply Chain Architecture and Personnel Management. The import of this is that government interference needs to be removed if refineries are to work optimally and the remaining four clusters should also be looked at in order to tackle these bottlenecks.