Distance-based Approach Research Articles

Minimal Learning Machine: A novel supervised distance-based approach for regression and classification

In this work, a novel supervised learning method, the Minimal Learning Machine (MLM), is proposed. Learning in MLM consists in building a linear mapping between input and output distance matrices. In the generalization phase, the learned distance map is used to provide an estimate of the distance from K output reference points to the unknown target output value. Then, the output estimation is formulated as multilateration problem based on the predicted output distance and the locations of the reference points. Given its general formulation, the Minimal Learning Machine is inherently capable of operating on nonlinear regression problems as well as on multidimensional response spaces. In addition, an intuitive extension of the MLM is proposed to deal with classification problems. A comprehensive set of computer experiments illustrates that the proposed method achieves accuracies that are comparable to more traditional machine learning methods for regression and classification thus offering a computationally valid alternative to such approaches.

Time series distance-based methods for non-intrusive load monitoring in residential buildings

Non-intrusive load monitoring (NILM) deals with the disaggregation of individual appliances from the total load at the smart meter level. This work proposes a generic methodology using temporal sequence classification algorithms. It is based on a low sampling rate unlike other approaches in this domain. An innovative time series distance-based approach in the temporal classification domain is compared with a standard NILM application based on the hidden Markov model (HMM) algorithm. The method is validated over a data-set of 100 houses for a duration of 1 year (with a 10min sampling rate). A qualitative analysis of the database is also conducted, allowing to segment it into four major clusters based on discussed features.

Localized Knowledge Spillovers and Patent Citations: A Distance-Based Approach

We develop a new distance-based test of localized knowledge spillovers that embeds the concept of control patents. Using microgeographic data, we identify localization distance for each technology class while allowing for spillovers across geographic units. We revisit the debate between Thompson and Fox-Kean (2005a, 2005b) and Henderson, Jaffe, and Trajtenberg (2005) on the existence of localized knowledge spillovers and find solid evidence supporting localization even when using finely grained controls. Unless biases induced by imperfect matching between citing and control patents due to unobserved heterogeneity are extremely large, our distance-based test detects localization for the majority of technology classes.

Object tracking with enhanced data association using a 3D range sensor for an unmanned ground vehicle

This paper describes a novel object tracking method which combines a conventional global nearest-neighbor (GNN) method with an enhanced data association approach using 3D data features for an unmanned ground vehicle. If an object crosses in front of an observation system which has a measurement sensor equipped on a vehicle, a conventional distance-based object tracking mechanism encounters a data association problem owing to high uncertainties and a leak of information: correspondence between observations and tracks. To overcome these correspondence problems, we use the object’s 3D information such as statistical and geometrical information. In this paper, we describe a clustering and segmentation processes in conjunction with a distance-based nearest-neighbor approach, RBNN-2, and applied GNN with enhanced data association using both a kinematic and geometric models derived from 3D range data. To evaluate our method using a 3D LADAR sensor, a Velodyne HDL-32e, installed on the vehicle, we conducted an experiment involving two crossing humans and a ground vehicle to identify different human trajectories for a leader-following system.

Fault diagnosis of rotating machinery based on kernel density estimation and Kullback-Leibler divergence

Based on kernel density estimation (KDE) and Kullback-Leibler divergence (KLID), a new data-driven fault diagnosis method is proposed from a statistical perspective. The ensemble empirical mode decomposition (EEMD) together with the Hilbert transform is employed to extract 95 time- and frequency-domain features from raw and processed signals. The distance-based evaluation approach is used to select a subset of fault-sensitive features by removing the irrelevant features. By utilizing the KDE, the statistical distribution of selected features can be readily estimated without assuming any parametric family of distributions; whereas the KLID is able to quantify the discrepancy between two probability distributions of a selected feature before and after adding a testing sample. An integrated Kullback- Leibler divergence, which aggregates the KLID of all the selected features, is introduced to discriminate various fault modes/damage levels. The effectiveness of the proposed method is demonstrated via the case studies of fault diagnosis for bevel gears and rolling element bearings, respectively. The observations from the case studies show that the proposed method outperforms the support vector machine (SVM)-based and neural network-based fault diagnosis methods in terms of classification accuracy. Additionally, the influences of the number of selected features and the training sample size on the classification performance are examined by a set of comparative studies.

Reduction in the Number of Comparisons Required to Create Matrix of Expert Judgment in the Comet Method

Abstract Multi-criteria decision-making (MCDM) methods are associated with the ranking of alternatives based on expert judgments made using a number of criteria. In the MCDM field, the distance-based approach is one popular method for receiving a final ranking. One of the newest MCDM method, which uses the distance-based approach, is the Characteristic Objects Method (COMET). In this method, the preferences of each alternative are obtained on the basis of the distance from the nearest characteristic ob jects and their values. For this purpose, the domain and fuzzy numbers set for all the considered criteria are determined. The characteristic objects are obtained as the combination of the crisp values of all the fuzzy numbers. The preference values of all the characteristic ob ject are determined based on the tournament method and the principle of indifference. Finally, the fuzzy model is constructed and is used to calculate preference values of the alternatives. In this way, a multi-criteria model is created and it is free of rank reversal phenomenon. In this approach, the matrix of expert judgment is necessary to create. For this purpose, an expert has to compare all the characteristic ob jects with each other. The number of necessary comparisons depends squarely to the number of ob jects. This study proposes the improvement of the COMET method by using the transitivity of pairwise comparisons. Three numerical examples are used to illustrate the efficiency of the proposed improvement with respect to results from the original approach. The proposed improvement reduces significantly the number of necessary comparisons to create the matrix of expert judgment.

Hesitant fuzzy QUALIFLEX approach with a signed distance-based comparison method for multiple criteria decision analysis

QUALIFLEX (QUALItative FLEXible multiple criteria method) is a very useful outranking method to deal simultaneously with the cardinal and ordinal information in decision making process. The purpose of this paper is to develop a hesitant fuzzy QUALIFLEX with a signed distance-based comparison method for handling multi-criteria decision-making problems in which both the assessments of alternatives on criteria and the weights of criteria are expressed by hesitant fuzzy elements (HFEs). We propose a novel concept of hesitancy index for the HFE to measure the degree of hesitancy of the decision-maker or the decision organization. By taking their hesitancy indices into account, we present a signed distance-based method to compare the magnitude of HFEs. Using the signed distance-based comparison approach, we define the concordance/discordance index, the weighted concordance/discordance index and the comprehensive concordance/discordance index. By investigating all possible permutations of alternatives with respect to the level of concordance/discordance of the complete preference order, the optimal ranking orders of alternatives can be obtained. An application study of the proposed method on green supplier selection is conducted. The study indicates that the proposed method does not require the complicated computation procedures but still yields a reasonable and credible solution.

GBAGC

Graph clustering, also known as community detection, is a long-standing problem in data mining. In recent years, with the proliferation of rich attribute information available for objects in real-world graphs, how to leverage not only structural but also attribute information for clustering attributed graphs becomes a new challenge. Most existing works took a distance-based approach. They proposed various distance measures to fuse structural and attribute information and then applied standard techniques for graph clustering based on these distance measures. In this article, we take an alternative view and propose a novel Bayesian framework for attributed graph clustering. Our framework provides a general and principled solution to modeling both the structural and the attribute aspects of a graph. It avoids the artificial design of a distance measure in existing methods and, furthermore, can seamlessly handle graphs with different types of edges and vertex attributes. We develop an efficient variational method for graph clustering under this framework and derive two concrete algorithms for clustering unweighted and weighted attributed graphs. Experimental results on large real-world datasets show that our algorithms significantly outperform the state-of-the-art distance-based method, in terms of both effectiveness and efficiency.

Identification of genetic variants associated with alternative splicing using sQTLseekeR.

Identification of genetic variants affecting splicing in RNA sequencing population studies is still in its infancy. Splicing phenotype is more complex than gene expression and ought to be treated as a multivariate phenotype to be recapitulated completely. Here we represent the splicing pattern of a gene as the distribution of the relative abundances of a gene’s alternative transcript isoforms. We develop a statistical framework that uses a distance-based approach to compute the variability of splicing ratios across observations, and a non-parametric analogue to multivariate analysis of variance. We implement this approach in the R package sQTLseekeR and use it to analyze RNA-Seq data from the Geuvadis project in 465 individuals. We identify hundreds of single nucleotide polymorphisms (SNPs) as splicing QTLs (sQTLs), including some falling in genome-wide association study SNPs. By developing the appropriate metrics, we show that sQTLseekeR compares favorably with existing methods that rely on univariate approaches, predicting variants that behave as expected from mutations affecting splicing.

Distance-based analysis of variance: Approximate inference

In several modern applications, ranging from genomics to neuroimaging, there is a need to compare measurements across different populations, such as those collected from samples of healthy and diseased individuals. The interest is in detecting a group effect, and typically many thousands or even millions of tests need to be performed simultaneously, as exemplified in genomics where single tests are applied for each gene across the genome. Traditional procedures, such as multivariate analysis of variance (MANOVA), are not suitable when dealing with nonvector-valued data structures such as functional or graph-structured observations. In this article, we discuss an existing distance-based MANOVA-like approach, the distance-based F (DBF) test, for detecting such differences. The null sampling distribution of the DBF test statistic relies on the distribution of the measurements and the chosen distance measure, and is generally unavailable in closed form. In practice, Monte Carlo permutation methods are deployed which introduce errors in estimating small p-values and increase familywise type I error rates when not using enough permutations. In this work, we propose an approximate distribution for the DBF test allowing inferences to be drawn without the need for costly permutations. This is achieved by approximating the permutation distribution that would be obtained by enumerating all permutations by the Pearson type III distribution using moment matching. The use of the Pearson type III distribution is motivated via empirical observations with real data. We provide evidence with real and simulated data that the resulting approximate null distribution of the DBF test is flexible enough to work well with a range of distance measures. Through extensive simulations involving different data types and distance measures, we provide evidence that the proposed methodology yields the same statistical power that would otherwise only be achievable if many millions of Monte Carlo permutations were performed.

What barcode sequencing reveals about the shark fishery in Peru

Elasmobranchs are rapidly declining due to overfishing and bycatch, underlining the need for immediate protection. Critical baseline information on the diversity of targeted species is, however, often missing. Peru is a major country for shark fishery, an activity that has been under-regulated and poorly monitored, aggravated by the superficial taxonomic identifications at landing points across the country. Furthermore, most of the species landed by the shark fishery in Peru are listed as Vulnerable in the IUCN Red List. To assess the diversity of shark species targeted by fisheries in Peru we analyzed the variation of the cytochrome oxidase I (cox1) region of the mitochondrial DNA from 118 samples collected between 2004 and 2009, from six landing points. Our analysis revealed unambiguously that the 16 shark species classified by fishermen using meristic characters corresponded only to nine species. While some commonly landed species (e.g. Prionace glauca) were consistently correctly identified, for others species multiple inconsistent names were applied (e.g. Galeorhinus galeus). Our molecular characterization further allowed the identification of specimens with non-informative common names (i.e. “tiburon”=shark). In most cases the unknown specimens were Isurus oxyrinchus and P. glauca. Interestingly, all samples labeled as common thresher (Alopias vulpinus) were identified as pelagic thresher (Alopias pelagicus). Finally, one sample was equivocally identified as a dusky shark (Carcharhinus obscurus) and as a galapagos shark (Carcharhinus galapagensis) reinforcing the genetic similarity reported for these species. We generated a character-based identification library containing 26 of the 31 commercially important sharks landed in Peru and tested its performance as a species diagnostic. The library correctly identified 25 out of 28 barcodes tested, outperforming the distance-based approach. This is the first study sequencing barcodes of marine species in Peru and generated a genetic reference library of targeted shark species. We suggest that the molecular tools used are a quick and effective complement for the monitoring of the fishery of threatened shark species. A combined effort to obtain these data, by countries in the east Pacific region with an on-going shark fishery, would provide with the essential guiding information to promote the implementation of effective sustainable management plans.

Bayesian Multiscale Modeling of Spatial Infrastructure Performance Predictions with an Application to Electric Power Outage Forecasting

A number of models have been developed to estimate the spatial distribution of infrastructure impact during a natural hazard event. For example, statistical approaches have been developed to estimate the percentage of customers without power attributable to a hurricane, with the estimates made at a local geography level such as census tracts. Whereas some statistical infrastructure performance models use extensive covariate data that captures a significant amount of the spatial information, others use a limited number of covariates to enhance model simplicity and to reduce the cost and time associated with obtaining these data. However, in these simpler models, the omitted covariates result in loss of spatial information in the model, leading to a situation in which predictions from adjacent regions are more dissimilar than would be expected. In this paper, a tree-based statistical mass-balance multiscale model is developed to smooth the outage predictions at granular levels by allowing spatially similar areas to inform one another with the goals of: (1) reducing spatial error in simplified prediction models, and (2) yielding estimates at other levels of aggregation in addition to the native model resolution. A generalized density-based clustering algorithm is used to extract the hierarchical spatial structure. The noise regions (i.e., those regions located in sparse areas) are then aggregated using a distance-based clustering approach. The authors demonstrate this approach using outage predictions from Hurricane Ivan and develop outage prediction maps at different levels of granularity. Language: en

Comments on “A distance-based statistical analysis of fuzzy number-valued data” by the SMIRE research group

This paper is a fine review of various aspects related to the statistical handling of “ontic” random fuzzy sets by the means of appropriate distances. It is quite comprehensive and helpful, as it clarifies the status of fuzzy sets in such methods, explains the advantages of using a distance-based approach, specifies the pitfalls in which one should not fall when dealing with “ontic” random fuzzy sets and provides some illustration of practical computations. Not being a statistician but an occasional user of statistics, my discussion will mainly focus on this more practical aspect.

A distance-based aggregation approach for group decision making with interval preference orderings

Xu (2013) proposed a nonlinear programming model to derive an exact formula to determine the experts’ relative importance weights for the group decision making (GDM) with interval preference orderings. However, in this study, we show that the exact formula to determine the weight vector which always equals to w=(1/m, 1/m,…,1/m)T (m is the number of experts). In this paper, we propose a distance-based aggregation approach to assess the relative importance weights for GDM with interval preference orderings. Relevant theorems are offered to support the proposed approach. After that, by using the weighted arithmetic averaging operator, we obtain the aggregated virtual interval preference orderings. We propose a possibility degree formula to compare two virtual interval preference orderings, then rank and select the alternatives. The proposed method is tested by two numerical examples. Comparative analysis are provided to show the advantages and effectiveness of the proposed method.

Robust outlier detection using the instability factor

Since outlier detection is applicable to various fields such as the financial, telecommunications, medical, and commercial industries, its importance is radically increasing. Receiving such great attention has led to the development of many detection methods, most of which pertain to either the distance-based approach or the density-based approach. However, each approach has intrinsic weaknesses. The former hardly detects local outliers, while the latter has the low density patterns problem. To overcome these weaknesses, we proposed a new detection method that introduces the instability factor of a data point by utilizing the concept of the center of gravity. The proposed method can be flexibly used for both local and global detection of outliers by controlling its parameter. In addition, it offers the instability plot containing useful information about the number and size of clusters in data. Numerical experiments based on artificial and real datasets show the effectiveness of the proposed method.

The Clustering of Categorical Data: A Comparison of a Model-based and a Distance-based Approach

For clustering multivariate categorical data, a latent class model-based approach (LCC) with local independence is compared with a distance-based approach, namely partitioning around medoids (PAM). A comprehensive simulation study was evaluated by both a model-based as well as a distance-based criterion. LCC was better according to the model-based criterion and PAM was sometimes better according to the distance-based criterion. However, LCC had an overall good and sometimes better distance-based performance as PAM, although this was not the case in a real data set on tribal art items.

Optimisation of cooperative spectrum sensing with ultra-wideband signalling

Reliable detection of primary user activity by spectrum sensing is of fundamental importance for cognitive radio systems. The objective of cooperative spectrum sensing is to combine the detection results of multiple cognitive radios in order to maximise the probability of detecting unused spectrum while meeting a required reliability of detecting primary user activity. In this paper, an optimisation strategy for cooperative spectrum sensing is proposed consisting of two parts. The first part is a Kullback-Leibler distance-based optimisation approach for the local decision thresholds. For the transmission of the local decisions, which can be realised by ultra-wideband underlay communications, in the second part an application-specific power control strategy is presented. The feasibility and effectiveness of the concepts are validated by real spectrum sensing results. The employed practical setup is based on software defined radio and detects a WiMAX-like OFDM signal.

On the Structure of European Economic and Monetary Union – Clues from the Data and Economic Intuition

In this work we aim to shed light on the dynamics of the convergence/divergence macroeconomic indicators of the EU-15 countries. The empirical distance-based approach is applied to evaluate the differences of the member states and localize the core-periphery issues of unionization.

Importance of Native Grassland Habitat for Den-Site Selection of Indian Foxes in a Fragmented Landscape

Fragmentation of native habitats is now a ubiquitous phenomenon affecting wildlife at various scales. We examined selection of den-sites (n = 26) by Indian foxes (Vulpes bengalensis) in a highly modified short-grassland landscape in central India (Jan-May, 2010). At the scale of the home-range, defined by an 800 m circular buffer around den sites, we examined the effect of land-cover edges and roads on selection of sites for denning using a distance-based approach. At the smaller den-area scale, defined by a 25 m x 25 m plot around den and paired available sites, the effect of microhabitat characteristics was examined using discrete-choice models. Indian foxes selected den-sites closer to native grasslands (t = -9.57, P < 0.001) and roads (t = -2.04, P = 0.05) than random at the home-range scale. At the smaller scale, abundance of rodents and higher visibility increased the odds of selection of a site by eight and four times respectively, indicating resource availability and predator avoidance to be important considerations for foxes. Indian foxes largely chose to den in human-made structures, indicated by the proportion of dens found in earthen bunds (0.69) and boulder piles (0.27) in the study area. With agricultural expansion and human modification threatening native short-grassland habitats, their conservation and effective management in human-dominated landscapes will benefit the Indian fox. The presence of some human-made structures within native grasslands would also be beneficial for this den-dependent species. We suggest future studies examine the impact of fragmentation and connectivity of grasslands on survival and reproductive success of the Indian fox.

Integration of data mining and multi-objective optimisation for decision support in production systems development

Multi-objective optimisation (MOO) is a powerful approach for generating a set of optimal trade-off (Pareto) design alternatives that the decision-maker can evaluate and then choose the most-suitable configuration, based on some high-level strategic information. Nevertheless, in practice, choosing among a large number of solutions on the Pareto front is often a daunting task, if proper analysis and visualisation techniques are not applied. Recent research advancements have shown the advantages of using data mining techniques to automate the post-optimality analysis of Pareto-optimal solutions for engineering design problems. Nonetheless, it is argued that the existing approaches are inadequate for generating high-quality results, when the set of the Pareto solutions is relatively small and the solutions close to the Pareto front have almost the same attributes as the Pareto-optimal solutions, of which both are commonly found in many real-world system problems. The aim of this paper is therefore to propose a distance-based data mining approach for the solution sets generated from simulation-based optimisation, in order to address these issues. Such an integrated data mining and MOO procedure is illustrated with the results of an industrial cost optimisation case study. Particular emphasis is paid to showing how the proposed procedure can be used to assist decision-makers in analysing and visualising the attributes of the design alternatives in different regions of the objective space, so that informed decisions can be made in production systems development.