Determinantal Clustering Processes - A Nonparametric Bayesian Approach to Kernel Based Semi-Supervised Clustering

Semi-supervised clustering is the task of clustering data points into clusters where only a fraction of the points are labelled. The true number of clusters in the data is often unknown and most models require this parameter as an input. Dirichlet process mixture models are appealing as they can infer the number of clusters from the data. However, these models do not deal with high dimensional data well and can encounter difficulties in inference. We present a novel nonparameteric Bayesian kernel based method to cluster data points without the need to prespecify the number of clusters or to model complicated densities from which data points are assumed to be generated from. The key insight is to use determinants of submatrices of a kernel matrix as a measure of how close together a set of points are. We explore some theoretical properties of the model and derive a natural Gibbs based algorithm with MCMC hyperparameter learning. The model is implemented on a variety of synthetic and real world data sets.

The influence of nitrogen clustering effect on optical transitions in GaInNAs/GaAs quantum wells

Photoluminescence (PL) spectra of GaInNAs/GaAs multiple quantum wells and GaInNAs epilayers grown on GaAs substrate show an apparent "S-shape" temperature-dependence of the of dominant luminescence peak. At low temperature and weak excitation conditions, a PL peak related to nitrogen cluster-induced bound states can be well resolved in the PL spectra. It displays a remarkable red shift of up to 60 meV and is thermally quenched below 100 K with increasing temperature, being attributed to N-cluster induced bound states. The indium incorporation exhibits significant effect on the cluster formation. The rapid thermal annealing treatment at 750 C can essentially remove the bound states-induced peak.

missing data imputation: a fuzzy k-means clustering algorithm over sliding window

Tianjin University of Technology

Confined intra-molecular clustering in orientated polyethylene after annealing

Deuterated polyethylene tracer molecules with small amount of branches (12 C2H5- branches per 1000 backbone carbon atoms) were blended with a hydrogenated polyethylene matrix to form a homogenous mixture. The conformational evolution of the deuterated chains in a stretched semi-cry stall me film was observed via online small angle neutron scattering measurements during annealing at high temperatures close to the melting point. Because the sample was annealed at a temperature closely below its melting point, the crystalline lamellae were only partially molten and the system could not fully relax. The global chain dimensions were preserved during annealing. Recrystallization of released polymeric chain segments allows for local phase separation thus driving the deuterated chain segments into the confining interlamellar amorphous layers giving rise to an interesting intra-molecular clustering effect of the long deuterated chain. This clustering is deduced from characteristic small angle neutron scattering patterns. The confined phase separation has its origin in primarily the small amount of the branches on the deuterated polymers which impede the crystallization of the deuterated chain segments.

Prediction of marine sediment corrosion by fuzzy clustering analysis

Along with the development of marine industries, especially marine petroleum exploitation, more and more pipelines are buried in the marine sediment. It is necessary and useful to know the corrosion environment and corrosiveness of marine sediment. In this paper, field corrosion environmental factors were investigated in Liaodong Bay marine sediment containing sulfate-reducing bacteria (SRB) and corrosion rate of steel in the partly sediment specimens were determined by the transplanting burying method. Based on the data, the fuzzy clustering analysis (FCA) was applied to evaluate and predict the corrosiveness of marine sediment. On that basis, the influence factors of corrosion damage were discussed.

Hierarchical Multi-classification with Predictive Clustering Trees in Functional Genomics

Struyf, J., Dzeroski, S. Blockeel, H. and Clare, A. (2005) Hierarchical Multi-classification with Predictive Clustering Trees in Functional Genomics. In proceedings of the EPIA 2005 CMB Workshop

Extraction and Clustering of Motion Trajectories in Video

A system is described that tracks moving objects in a video dataset so as to extract a representation of the objects' 3D trajectories. The system then finds hierarchical clusters of similar trajectories in the video dataset. Objects' motion trajectories are extracted via an EKF formulation that provides each object's 3D trajectory up to a constant factor. To increase accuracy when occlusions occur, multiple tracking hypotheses are followed. For trajectory-based clustering and retrieval, a modified version of edit distance, called longest common subsequence (LCSS) is employed. Similarities are computed between projections of trajectories on coordinate axes. Trajectories are grouped based, using an agglomerative clustering algorithm. To check the validity of the approach, experiments using real data were performed.

DNS-based Internet Client Clustering and Characterization

This paper proposes a novel protocol which uses the Internet Domain Name System (DNS) to partition Web clients into disjoint sets, each of which is associated with a single DNS server. We define an L-DNS cluster to be a grouping of Web Clients that use the same Local DNS server to resolve Internet host names. We identify such clusters in real-time using data obtained from a Web Server in conjunction with that server's Authoritative DNS―both instrumented with an implementation of our clustering algorithm. Using these clusters, we perform measurements from four distinct Internet locations. Our results show that L-DNS clustering enables a better estimation of proximity of a Web Client to a Web Server than previously proposed techniques. Thus, in a Content Distribution Network, a DNS-based scheme that redirects a request from a web client to one of many servers based on the client's name server coordinates (e.g., hops/latency/loss-rates between the client and servers) would perform better with our algorithm.