Application Of Spectral Clustering Research Articles

Triangular fuzzy-based spectral clustering for energy-efficient routing in wireless sensor network

In this study, a triangular fuzzy-based spectral cluster routing (TF-SCR) mechanism is proposed in order to improve network lifetime and reliability of data packet transmission in WSN with minimum routing overhead. The designed TF-SCR mechanism contains three steps, namely preprocessing, clustering and post-processing. TF-SCR mechanism initially carried out preprocessing where the residual energy and received signal strength of all sensor nodes in WSN is determined. After preprocessing, TF-SCR mechanism performs clustering process with the application of spectral clustering in which the sensor nodes in WSN are grouped based on residual energy level. After completing clustering process, TF-SCR mechanism implements post-processing step where triangular fuzzy membership function is applied to select the sensor node with higher residual energy and signal strength as cluster head for routing data packets to sink nodes in WSN. This helps for TF-SCR mechanism to obtain energy-efficient routing in WSN. Thus, TF-SCR mechanism enhances the network lifetime and reliability of data broadcasting in WSN as compared to state-of-the-art works. The simulation of TF-SCR mechanism is conducted on factors such as energy consumption, network lifetime, routing overhead and reliability with respect to a number of sensor nodes and data packets. The simulation results demonstrate that the TF-SCR mechanism is able to improve the network lifetime and also minimize the overhead of routing with minimal energy consumption as compared to state-of-the-art works.

Spatial Data Mining with the Application of Spectral Clustering: A Trend Detection Approach

Spatial Data Mining with the Application of Spectral Clustering: A Trend Detection Approach

Temporal dietary patterns derived from Spectral clustering and factor analysis are associated with diet quality using NHANES 1999‐2004.

ObjectiveThe study evaluated the relationship of temporal dietary patterns derived from a novel application of spectral clustering and factor analysis, to dietary quality among adult participants from the National Health and Nutrition Examination Survey (NHANES, 1999‐2004).MethodsNHANES adult participants (20‐65y) with a complete day‐one 24‐hour dietary recall were clustered based on proportional energy intake, time of eating, and number of eating occasions, using algorithms of spectral clustering analysis and factor analysis. The association between clusters and mean dietary quality, as indicated by the 2005 Healthy Eating Index (HEI), was evaluated using multiple linear regression and controlled for potential confounders.ResultsMean dietary quality among the four clusters representing distinct temporal dietary patterns was significantly different (P<0.0001). The cluster representing a pattern with proportionally equivalent energy intake consumed at three evenly spaced intake occasions throughout the daytime had the highest HEI score (50.1±0.3, mean±SEM).ImplicationKnowledge of the temporal aspect of dietary patterns may enhance understanding of the relationship of diet to health within the context of daily patterns.FundingPurdue University

Spectral Clustering of Microarray Data Elucidates the Roles of Microenvironment Remodeling and Immune Responses in Survival of Head and Neck Squamous Cell Carcinoma

To identify functionally related prognostic gene sets for head and neck squamous cell carcinoma (HNSCC) by unsupervised statistical analysis of microarray data. Microarray analysis was performed on 14 normal oral epithelium and 71 HNSCCs from patients with outcome data. Spectral clustering (SC) analysis of the data set identified multiple vectors representing distinct aspects of gene expression heterogeneity between samples. Gene ontology (GO) analysis of vector gene lists identified gene sets significantly enriched within defined biologic pathways. The prognostic significance of these was established by Cox survival analysis. The most influential SC vectors were V2 and V3. V2 separated normal from tumor samples. GO analysis of V2 gene lists identified pathways with heterogeneous expression between HNSCCs, notably focal adhesion (FA)/extracellular matrix remodeling and cytokine-cytokine receptor (CR) interactions. Similar analysis of V3 gene lists identified further heterogeneity in CR pathways. V2CR genes represent an innate immune response, whereas high expression of V3CR genes represented an adaptive immune response that was not dependent on human papillomavirus status. Survival analysis demonstrated that the FA gene set was prognostic of poor outcome, whereas classification for adaptive immune response by the CR gene set was prognostic of good outcome. A combined FA&CR model dramatically exceeded the performance of current clinical classifiers (P < .001 in our cohort and, importantly, P = .007 in an independent cohort of 60 HNSCCs). The application of SC and GO algorithms to HNSCC microarray data identified gene sets highly significant for predicting patient outcome. Further large-scale studies will establish the usefulness of these gene sets in the clinical management of HNSCC.

Spectral clustering in multi-agent systems

We examine the application of spectral clustering for breaking up the behavior of a multi-agent system in space and time into smaller, independent elements. We propose clustering observations of individual entities in order to identify significant changes in the parameter space (like spatial position) and detect temporal alterations of behavior within the same framework. Available knowledge of important interactions (events) between entities is also considered. We describe a novel algorithm utilizing iterative subdivisions where clusters are pre-processed at each step to counter spatial scaling, rotation, replay speed, and varying sampling frequency. A method is presented to balance spatial and temporal segmentation based on the expected group size, and a validity measure is introduced to determine the optimal number of clusters. We demonstrate our results by analyzing the outcomes of computer games and compare our algorithm to K-means and traditional spectral clustering.