Hierarchical Gaussian Mixture Model with Objects Attached to Terminal and Non-terminal Dendrogram Nodes
نویسندگان
چکیده
A hierarchical clustering algorithm based on Gaussian mixture model is presented. The key difference to regular hierarchical mixture models is the ability to store objects in both terminal and nonterminal nodes. Upper levels of the hierarchy contain sparsely distributed objects, while lower levels contain densely represented ones. As it was shown by experiments, this ability helps in noise detection (modelling). Furthermore, compared to regular hierarchical mixture model, the presented method generates more compact dendrograms with higher quality measured by adopted F-measure.
منابع مشابه
On the Achievable Rate-Regions for the Gaussian Two-way Diamond Channels
In this channel,we study rate region of a Gaussian two-way diamond channel which operates in half-duplex mode. In this channel, two transceiver (TR) nodes exchange their messages with the help of two relay nodes. We consider a special case of the Gaussian two-way diamond channels which is called Compute-and-Forward Multiple Access Channel (CF-MAC). In the CF-MAC, the TR nodes transmit their mes...
متن کاملRobust Method for E-Maximization and Hierarchical Clustering of Image Classification
We developed a new semi-supervised EM-like algorithm that is given the set of objects present in eachtraining image, but does not know which regions correspond to which objects. We have tested thealgorithm on a dataset of 860 hand-labeled color images using only color and texture features, and theresults show that our EM variant is able to break the symmetry in the initial solution. We compared...
متن کاملTrajectory tracking of under-actuated nonlinear dynamic robots: Adaptive fuzzy hierarchical terminal sliding-mode control
In recent years, underactuated nonlinear dynamic systems trajectory tracking, such as space robots and manipulators with structural flexibility, has become a major field of interest due to the complexity and high computational load of these systems. Hierarchical sliding mode control has been investigated recently for these systems; however, the instability phenomena will possibly occur, especia...
متن کاملArtifcial neural network approach for the prediction of terminal falling velocity of non-spherical particles through Newtonian and non-Newtonian fluids
The investigation of the terminal falling velocity of non-spherical particles is currently one of the most promising topics in sedimentation technology due to its great signifcance in many separation processes. In this study, the potential of Artifcial Neural Networks (ANNs) for the prediction of nonspherical particles terminal falling velocity through Newtonian and nonNewtonian (power law) liq...
متن کاملFAST ATOM BOMBARDMENT MASS SPECTROMETRY (FABMS) ANALYSIS OF AN N- TERMINAL - BLOCKED PEPTIDE
FABMS analysis of T-lb peptide before and after one cycle of Edman degradation indicated an unblocked N-terminal Thr residue for this tryptic peptide. In contrast , our data showed a molecular protonated ion, MH + for T- la peptide at 655 mass units (mu) which is 42 mu higher than the MH ion of T- 1b peptide. In addition, T- la peptide was not amenable to one cycle of manual Edman degrada...
متن کامل