Kth-nearest Neighbor Research Articles

Use of Laser Capture Microdissection to Map Hepatitis C Virus–Positive Hepatocytes in Human Liver

Hepatitis C virus (HCV) predominantly infects hepatocytes, but many hepatocytes are not infected; studies have shown that HCV antigens cluster within the liver. We investigated spatial distribution and determinants of HCV replication in human liver samples. We analyzed liver samples from 4 patients with chronic HCV infection (genotype 1, Metavir scores 0-1) to estimate the proportion of infected hepatocytes and the amount of HCV viral RNA (vRNA) per cell. Single-cell laser capture microdissection was used to capture more than 1000 hepatocytes in grids, to preserve geometric relationships. HCV vRNA and interferon-induced transmembrane protein 3 (IFITM3) messenger RNA (the transcript of an interferon-stimulated gene) were measured in the same hepatocytes by quantitative polymerase chain reaction and assembled in maps to identify areas of high and low HCV replication. Patients' serum levels of HCV RNA ranged from 6.87 to 7.40 log10 IU/mL; the proportion of HCV-infected hepatocytes per person ranged from 21% to 45%, and the level of vRNA ranged from 1 to 50 IU/hepatocyte. Infection was not random; we identified clustering of HCV-positive hepatocytes using infected-neighbor analysis (P < .0005) and distance to the kth nearest neighbor compared with random distributions, obtained by bootstrap simulations (P < .02). Hepatocytes that expressed IFITM3 did not appear to cluster and were largely HCV negative. We used single-cell laser capture and high-resolution analysis to show that in human liver HCV infects hepatocytes in nonrandom clusters, whereas expression of antiviral molecules is scattered among hepatocytes. These findings show that quantitative single-cell RNA measurements can be used to estimate the abundance of HCV vRNA per infected human hepatocyte and are consistent with cell-cell propagation of infection in the absence of clustered IFITM3.

Neighborhood topological effect on grain topology-size relationship in three-dimensional polycrystalline microstructures

A new grain topology-size relationship in three-dimensional (3D) polycrystalline microstructures has recently been established by considering the effects of non-random first nearest neighbor grains. In this contribution, a generalized form for this relationship is presented by considering the interactions of kth (k=1, 2, 3…) nearest neighbor grains, and large scale Monte Carlo-Potts model simulation is used to investigate the general neighborhood topological effect on the topology-size relationship. The results show that, unlike their first nearest neighbors (k=1), the topological correlations of 3D grains with their kth layers (k⩾2) of nearest-neighbors may have trivial effect on the topology-size relationship.

Authentications of Myanmar National Registration Card

The automatic identification system of Myanmar national registration card (NRC) holder is presented in this paper. The proposed system can be handled the identification by the extracted low quality face image and fingerprint image from Myanmar NRC. Both of the facial recognition and fingerprint recognition system are developed for Myanmar citizenship confirmation. Age invariant face recognition algorithm is performed based on combination of DiaPCA (Diagonal principal Component Analysis) and KNN (K th nearest neighbor classifier) approaches. An algorithm of the fingerprint recognition is proposed for recognition of the poor quality fingerprint image with fabric background. Several experiments have been done for confirming the effectiveness of the proposed approach.

Power‐aware range‐free wireless sensor network localization using neighbor distance distribution

ABSTRACTIn large‐scale wireless sensor networks, cost‐effective and energy‐efficient localization of sensor nodes is an important research topic. In spite of their coarse accuracy, range‐free (connectivity‐based) localization methods are considered as cost‐effective alternatives to the range‐based localization schemes with specialized hardware requirements.In this paper, we derive closed‐form expressions for the average minimum transmit powers required for the localization of sensor nodes, under deterministic path loss, log‐normal shadowing, and Rayleigh fading channel models. The impacts of propagation environment and spatial density of anchor nodes on the minimum transmit power for node localization are evaluated analytically as well as through simulations. Knowledge of the minimum transmit power requirements for localizability of a sensor node enables improving energy efficiency and prolonging lifetime of the network. We also propose a novel distance metric for range‐free localization in large‐scale sensor networks. The target and anchor nodes are assumed to be positioned according to two statistically independent two‐dimensional homogeneous Poisson point processes. Analytical expression for the average distance from a target node to its kth nearest neighbor anchor node is derived and is used for estimating the target‐to‐anchor node distances for localization. The Cramér–Rao lower bound on the localization accuracy for the new distance estimator is derived. Simulation results show the accuracy of the proposed distance estimate compared with some existing ones for range‐free localization. The results of our investigation are significant for low‐cost, energy‐efficient localization of wireless sensor nodes. Copyright © 2011 John Wiley &amp; Sons, Ltd.

Nearest neighbour diagnostic statistics on the accuracy of APT solute cluster characterisation

Diagnostic statistics and information theory techniques have been developed to investigate the accuracy to which solute clusters characterised in atom probe tomography (APT) data can reflect the true nature of the physical clusters in the original specimen. Simulated atom-probe datasets representing a range of atomic solute clustering within a pseudo-binary alloy upon an fcc aluminium lattice were generated for the study. The effectiveness of partitioning the APT-like simulated data based upon a binary classification defined by a distance threshold d max upon the kth nearest neighbour distance distribution was investigated. Information theory was also used to optimise the selection of the threshold d max. Analysis of variation was performed upon a factorial design of data simulations with low and high levels of: solute concentration; short-range order; and background to the mass-to-charge-state-ratio spectrum. This meta-analysis showed that the background levels have a significant compromising effect upon the binary classification in low solute systems with relatively low or random levels of clustering. Although the random clustering of higher solute concentrations is better analysed, significantly non-random clustering in both low and high solute concentrations is analysed well despite the presence of high levels of background. A meta-analysis of the binary classification upon a simulated dispersion of coherent precipitates within a similar matrix was also undertaken. Optimal k and d max parameters are likely a dependent upon the physical dimensions of precipitate size as well as the precipitate/matrix solute concentrations.

Multi-scale decomposition of point process data

To automatically identify arbitrarily-shaped clusters in point data, a theory of point process decomposition based on kth Nearest Neighbour distance is proposed. We assume that a given set of point data is a mixture of homogeneous processes which can be separated according to their densities. Theoretically, the local density of a point is measured by its kth nearest distance. The theory is divided into three parts. First, an objective function of the kth nearest distance is constructed, where a point data set is modelled as a mixture of probability density functions (pdf) of different homogeneous processes. Second, we use two different methods to separate the mixture into different distinct pdfs, representing different homogeneous processes. One is the reversible jump Markov Chain Monte Carlo strategy, which simultaneously separates the data into distinct components. The other is the stepwise Expectation-Maximization algorithm, which divides the data progressively into distinct components. The clustering result is a binary tree in which each leaf represents a homogeneous process. Third, distinct clusters are generated from each homogeneous point process according to the density connectivity of the points. We use the Windowed Nearest Neighbour Expectation-Maximization (WNNEM) method to extend the theory and identify the spatiotemporal clusters. Our approach to point processes is similar to wavelet transformation in which any function can be seen as the summation of base wavelet functions. In our theory, any point process data set can be viewed as a mixture of a finite number of homogeneous point processes. The wavelet transform can decompose a function into components of different frequencies while our theory can separate point process data into homogeneous processes of different densities. Two experiments on synthetic data are provided to illustrate the theory. A case study on reservoir-induced earthquakes is also given to evaluate the theory. The results show the theory clearly reveals spatial point patterns of earthquakes in a reservoir area. The spatiotemporal relationship between the main earthquake and the clustered earthquake (namely, foreshocks and aftershocks) was also revealed.

ESTIMATION OF MULTIVARIATE SHANNON ENTROPY USING MOMENTS

Summary Three new entropy estimators of multivariate distributions are introduced. The two cases considered here concern when the distribution is supported by a unit sphere and by a unit cube. In the former case, the consistency and the upper bound of the absolute error for the proposed entropy estimator are established. In the latter one, under the assumption that only the moments of the underlying distribution are available, a non-traditional estimator of the entropy is suggested. We also study the practical performances of the constructed estimators through simulation studies and compare the estimators based on the moment-recovered approaches with their counterparts derived by using the histogram and kth nearest neighbour constructions. In addition, one worked example is briefly discussed.

Fuzzy C-Means Clustering of Signal Functional Principal Components for Post-Processing Dynamic Scenarios of a Nuclear Power Plant Digital Instrumentation and Control System

This paper addresses the issue of the classification of accident scenarios generated in a dynamic safety and reliability analyses of a Nuclear Power Plant (NPP) equipped with a Digital Instrumentation and Control system (I&C). More specifically, the classification of the final state reached by the system at the end of an accident scenario is performed by Fuzzy C-Means clustering the Functional Principal Components (FPCs) of selected relevant process variables. The approach allows capturing the characteristics of the process evolution determined by the occurrence, timing, and magnitudes of the fault events. An illustrative case study is considered, regarding the fault scenarios of the digital I&C system of the Lead Bismuth Eutectic eXperimental Accelerator Driven System (LBE-XADS). The results obtained are compared with those of the Kth Nearest Neighbor (KNN), and Classification and Regression Tree (CART) classifiers.

Magnetic Resonance

This paper presents a novel methodology for the non-rigid registration of cardiac perfusion MRI exams. The target medical application is the automated quantitative assessment of myocardial perfusion for clinical diagnosis and longitudinal study of ischemic pathologies. Specifically, an original variational method for the groupwise registration of p-MRI exams based on high-dimensional feature distribution matching using (normalized) mutual information, is developed. The hard issue of estimating information in high-dimensional spaces is solved using state-of-the-art kth-nearest neighbor (kNN) entropy estimators. Combined with mean-shift approximation, the latter allow to efficiently optimize (normalized) mutual information over finite- and infinite-dimensional motion spaces. This framework is applied to the groupwise alignment of cardiac p-MRI exams using local contrast enhancement curves as a feature set, and a B-spline model for cardio-thoracic motions. Preliminary experimental assessment suggests that the technique allows for accurately aligning up to 34 p-MRI images simultaneously, and for further reliably computing perfusion parameters whose joint analysis strongly correlates with expert-based visual diagnosis.

IMORC: An infrastructure and architecture template for implementing high-performance reconfigurable FPGA accelerators

The design, implementation and optimization of FPGA accelerators is a challenging task, especially when the accelerator comprises multiple compute cores distributed across CPU and FPGA resources and memories and exhibits data-dependent runtime behavior. In order to simplify the development of FPGA accelerators we propose IMORC, an infrastructure and architecture template that helps raising the level of abstraction. The IMORC development flow bases on a modeling technique for visualizing an application’s communication demand and an architecture template that aids the developer in implementing the design. The architectural template consists of a versatile on-chip interconnect with asynchronous FIFOs and bitwidth conversion placed into the communication links, a performance monitoring infrastructure for collecting performance information during runtime and a set of generic infrastructure cores which are frequently needed in accelerator designs. We demonstrate the usefulness of the IMORC development flow by means of the case study of accelerating the kth nearest neighbor thinning problem, where IMORC greatly helps us in understanding the communication demand and in implementing the application. With the integrated performance monitoring infrastructure, we gain insights into the data-dependent behavior of the accelerator that helps us in identifying bottlenecks and optimizing the accelerator to achieve a speedup of 10× to 40× over an optimized CPU implementation.

Correction: A class of Rényi information estimators for multidimensional densities

and Hq = ^(1 ? f^m fq(x)dx), q ^ 1, of an unknown probability measure with density / on Rm with respect to the Lebesgue measure though kth nearest neighbor distances in a sample X\,..., Xjy i.i.d. with the density /. The results in [1] about the asymptotic unbiasedness and consistency of the estimator proposed for Iq = E{fq~l(X)} = fRm fq(x)dx are correct for q > 1 but, for q 0: fmm \x\rf(x)dx rc(/)(see[2]). A correct version of the convergence results of lN,k,q to for 0 < g < 1 and / having unbounded support can be obtained by using results on the sub additivity of Euclidean functionals (see [3]) and can be formulated as follows [2]:

OP-KNN: Method and Applications

This paper presents a methodology named Optimally Pruned K-Nearest Neighbors (OP-KNNs) which has the advantage of competing with state-of-the-art methods while remaining fast. It builds a one hidden-layer feedforward neural network using K-Nearest Neighbors as kernels to perform regression. Multiresponse Sparse Regression (MRSR) is used in order to rank each kth nearest neighbor and finally Leave-One-Out estimation is used to select the optimal number of neighbors and to estimate the generalization performances. Since computational time of this method is small, this paper presents a strategy using OP-KNN to perform Variable Selection which is tested successfully on eight real-life data sets from different application fields. In summary, the most significant characteristic of this method is that it provides good performance and a comparatively simple model at extremely high-learning speed.

Detecting feature from spatial point processes using Collective Nearest Neighbor

In a spatial point set, clustering patterns (features) are difficult to locate due to the presence of noise. Previous methods, either using grid-based method or distance-based method to separate feature from noise, suffer from the parameter choice problem, which may produce different point patterns in terms of shape and area. This paper presents the Collective Nearest Neighbor method (CLNN) to identify features. CLNN assumes that in spatial data clustered points and noise can be viewed as two homogenous point processes. The one with the higher intensity is considered as a feature and the one with the lower intensity is treated as noise. As a result, they can be separated according to the difference in intensity between them. With CLNN, points are first classified into feature and noise based on the kth nearest distance (the distance between a point and its kth nearest neighbor) at various values of k. Then, CLNN selects those classifications in which the separated classes (i.e. features and noise) are homogenous Poisson processes and cannot be further divided. Finally, CLNN identifies clustered points by averaging the selected classifications. Evaluation of CLNN using simulated data shows that CLNN reduces the number of false points significantly. The comparison between CLNN, the shared nearest neighbor, the spatial scan and the classification entropy method shows that CLNN produced the fewest false points. A case study using seismic data in southwestern China showed that CLNN is able to identify foreshocks of the Songpan earthquake ( M = 7.2), which may help to locate the epicenter of the Songpan earthquake.

A Plotless Density Estimator Based on the Asymptotic Limit of Ordered Distance Estimation Values

Abstract Estimation of tree density from point-tree distances is an attractive option for quick inventory of new sites, but estimators that are unbiased in clustered and dispersed situations have not been found. Noting that bias of an estimator derived from distances to the kth nearest neighbor from a random point tends to decrease with increasing k, a method is proposed for estimating the limit of an asymptotic function through a set of ordered distance estimators. A standard asymptotic model is derived from the limiting case of a clustered distribution. The proposed estimator is evaluated against 13 types of simulated generating processes, including random, clustered, dispersed, and mixed. Performance is compared with ordered distance estimation of the same rank and with fixed-area sampling with the same number of trees tallied. The proposed estimator consistently performs better than ordered distance estimation and nearly as well as fixed-area sampling in all but the most clustered situations. The estimator also provides information regarding the degree of clustering or dispersion.

Order Distance in Regular Point Patterns

This article examines the kth nearest neighbor distance for three regular point patterns: square, triangular, and hexagonal lattices. The probability density functions of the kth nearest distance and the average kth nearest distances are theoretically derived for k=1, 2, …, 7. As an application of the kth nearest distance, we consider a facility location problem with closing of facilities. The problem is to find the optimal regular pattern that minimizes the average distance to the nearest open facility. Assuming that facilities are closed independently and at random, we show that the triangular lattice is optimal if at least 68% of facilities are open by comparing the upper and lower bounds of the average distances.

The application of SOM as a decision support tool to identify AACSB peer schools

For a business school, the selection of its peer schools is an important component of its International Association for Management Education (AACSB) (re)accreditation process. A school typically compares itself with other institutions having similar structural and identity-based attributes. The identification of peer schools is critical and can have a significant impact on a business school's accreditation efforts. For many schools the selection of comparable peer schools is a judgmental process. This study offers an alternative means for selection; a quantitative technique called Kohonen's Self-Organizing Map (SOM) network for clustering. In this research, we first demonstrate the capability of SOM as a clustering tool to visually uncover the relationships among AACSB-accredited schools. The results suggest that SOM is an effective and robust clustering method. Then, we compare the results of SOM with that of other clustering methods, such as K-means, Factor/ K-means analysis, and kth nearest neighbor procedure. The objective of this study is to demonstrate that a two-dimensional SOM map can be used to integrate the results of various clustering methods and, thus, act as a visual decision support tool.

A class of Rényi information estimators for multidimensional densities

A class of estimators of the Rényi and Tsallis entropies of an unknown distribution f in ℝm is presented. These estimators are based on the kth nearest-neighbor distances computed from a sample of N i.i.d. vectors with distribution f. We show that entropies of any order q, including Shannon’s entropy, can be estimated consistently with minimal assumptions on f. Moreover, we show that it is straightforward to extend the nearest-neighbor method to estimate the statistical distance between two distributions using one i.i.d. sample from each.

Choice of neighbor order in nearest-neighbor classification

The kth-nearest neighbor rule is arguably the simplest and most intuitively appealing nonparametric classification procedure. However, application of this method is inhibited by lack of knowledge about its properties, in particular, about the manner in which it is influenced by the value of k; and by the absence of techniques for empirical choice of k. In the present paper we detail the way in which the value of k determines the misclassification error. We consider two models, Poisson and Binomial, for the training samples. Under the first model, data are recorded in a Poisson stream and are “assigned” to one or other of the two populations in accordance with the prior probabilities. In particular, the total number of data in both training samples is a Poisson-distributed random variable. Under the Binomial model, however, the total number of data in the training samples is fixed, although again each data value is assigned in a random way. Although the values of risk and regret associated with the Poisson and Binomial models are different, they are asymptotically equivalent to first order, and also to the risks associated with kernel-based classifiers that are tailored to the case of two derivatives. These properties motivate new methods for choosing the value of k.

Estimating Relative Risk on the Line Using Nearest Neighbor Statistics

This paper considers a non-parametric method for identifying intervals on the line where the relative risk of cases to controls exceeds a pre-specified level. The method is based on the kth nearest neighbor (kNN) approach for density estimation. An asymptotic result is presented that yields an explicit formula for constructing a confidence interval for the relative risk at a given point. Numerical simulations are used to compare this approach with a kernel density estimation procedure. An application is made to a case-control study in which the relative risk of motor vehicle crashes caused by female drivers is compared to male drivers in the state of Kentucky as a function of age and then by time of day.

Application of wrapper approach and composite classifier to the stock trend prediction

The research on the stock market prediction has been more popular in recent years. Numerous researchers tried to predict the immediate future stock prices or indices based on technical indices with various mathematical models and machine learning techniques such as artificial neural networks (ANN), support vector machines (SVM) and ARIMA models. Although some researches in the literature exhibit satisfactory prediction achievement when the average percentage error and root mean square error are used as the performance metrics, the prediction accuracy of whether stock market goes or down is seldom analyzed. This paper employs wrapper approach to select the optimal feature subset from original feature set composed of 23 technical indices and then uses voting scheme that combines different classification algorithms to predict the trend in Korea and Taiwan stock markets. Experimental result shows that wrapper approach can achieve better performance than the commonly used feature filters, such as χ 2-Statistic, Information gain, ReliefF, Symmetrical uncertainty and CFS. Moreover, the proposed voting scheme outperforms single classifier such as SVM, kth nearest neighbor, back-propagation neural network, decision tree, and logistic regression.