Search

In the new "gig" economy, a user plays the role of a consumer as well as a service provider. As a service provider, drivers travelling from a source to a destination may opportunistically pickup and drop-off packages along the way if that does not add significantly to their trip distance or time. This gives rise to a new business offering called Package Delivery as a Service (PDaaS) that brokers package pickups and deliveries at one end and connects them to drivers on the other end, thus creating an ecosystem of supply and demand. The dramatic cost savings of such a service model come from the fact that the driver is already en-route to their destination and the package delivery adds a small overhead to an already pre-planned trip. From a technical perspective, this problem introduces new technical challenges that are uncommon in the literature. The driver may want to optimise for distance or time. Furthermore, new packages arrive for delivery all the time and are assigned to various drivers continuously. This means that the algorithm has to work in an environment that is dynamic, thereby precluding most standard road network precomputation efforts. Furthermore, the number of packages that are available for delivery could be in the hundreds of thousands, which has to be quickly pruned down for the algorithm to scale. The paper proposes a variation called dual Dijkstra's that combines a forward and a backward scan in order to find delivery options that satisfy the constraints specified by the driver. The new dual heuristic improves the standard single Dijkstra's approach by narrowing down the search space, thus resulting in significant speed-ups over the standard algorithms. Furthermore, a combination of dual Dijkstra's with a heuristic landmark approach results in a dramatic speed-up compared to the baseline algorithms. Experimental results show that the proposed approach can offer drivers a choice of packages to deliver under specified constraints of time or distance, and providing sub-second response time despite the complexity of the problem involved. As the number of packages in the system increases, the matchmaking process becomes easier resulting in faster response times. The scalability of the PDaaS infrastructure is demonstrated using extensive experimental results. [ABSTRACT FROM AUTHOR]

The paper aimed to examine the price development of gold, silver, and platinum from 1 January 2019 to 27 November 2023. The aim was also to predict the price development of pairs of metals - gold and silver, gold and platinum, and silver and platinum until the end of 2024. The Nearest Neighbours method in Wolfram Mathematica software was used to achieve the set goal. In terms of the development of metal pairs until the end of 2024, it was found that the most profitable pairs for investments are gold-platinum and silver-platinum. If an investor is to invest in a single metal, platinum is recommended. In the case of a pair of metals, it would be appropriate to choose gold-platinum or silver-platinum. The research is limited in terms of the metals selected, a narrower set of historical data to make a prediction, or the selected method. [ABSTRACT FROM AUTHOR]

We study the |$k$| -nearest neighbour classifier (⁠|$k$| -NN) of probability measures under the Wasserstein distance. We show that the |$k$| -NN classifier is not universally consistent on the space of measures supported in |$(0,1)$|⁠. As any Euclidean ball contains a copy of |$(0,1)$|⁠ , one should not expect to obtain universal consistency without some restriction on the base metric space, or the Wasserstein space itself. To this end, via the notion of |$\sigma $| -finite metric dimension, we show that the |$k$| -NN classifier is universally consistent on spaces of discrete measures (and more generally, |$\sigma $| -finite uniformly discrete measures) with rational mass. In addition, by studying the geodesic structures of the Wasserstein spaces for |$p=1$| and |$p=2$|⁠ , we show that the |$k$| -NN classifier is universally consistent on spaces of measures supported on a finite set, the space of Gaussian measures and spaces of measures with finite wavelet series densities. [ABSTRACT FROM AUTHOR]

Graphs are versatile structures for the representation of many real-world data. Deep Learning on graphs is currently able to solve a wide range of problems with excellent results. However, both the generation of graphs and the handling of large graphs still remain open challenges. This work aims to introduce techniques for generating large graphs and test the approach on a complex problem such as the calculation of dose distribution in oncological radiotherapy applications. To this end, we introduced a pooling technique (ReNN-Pool) capable of sampling nodes that are spatially uniform without computational requirements in both model training and inference. By construction, the ReNN-Pool also allows the definition of a symmetric un-pooling operation to recover the original dimensionality of the graphs. We also present a Variational AutoEncoder (VAE) for generating graphs, based on the defined pooling and un-pooling operations, which employs convolutional graph layers in both encoding and decoding phases. The performance of the model was tested on both the realistic use case of a cylindrical graph dataset for a radiotherapy application and the standard benchmark dataset sprite. Compared to other graph pooling techniques, ReNN-Pool proved to improve both performance and computational requirements. [ABSTRACT FROM AUTHOR]

Clustering on the data with multiple aspects, such as multi-view or multi-type relational data, has become popular in recent years due to their wide applicability. The approach using manifold learning with the Non-negative Matrix Factorization (NMF) framework, that learns the accurate low-rank representation of the multi-dimensional data, has shown effectiveness. We propose to include the inter-manifold in the NMF framework, utilizing the distance information of data points of different data types (or views) to learn the diverse manifold for data clustering. Empirical analysis reveals that the proposed method can find partial representations of various interrelated types and select useful features during clustering. Results on several datasets demonstrate that the proposed method outperforms the state-of-the-art multi-aspect data clustering methods in both accuracy and efficiency. [ABSTRACT FROM AUTHOR]

Advances in data acquisition have generated an enormous amount of data that captures business, commercial, technological and scientific information. However, some occurrences are rare or unusual, irrespective of a large amount of data available. These rare occurrences in data mining are usually referred to as outliers or anomalies. All these rare occurrences are infrequent. Sometimes it varies from 0.01% to 10% depending on the type of application. In recent years, outlier detection has become important in many applications and has attracted considerable attention among the increasing number of data mining techniques. Focusing on this has resulted in several outlier detection algorithms, mostly based on distance or density. However, each method has its inherent weaknesses. Methods based on distance have problems with local density, and methods based on density have problems with low-density patterns. In this paper, we present a new outlier detection algorithm based on the relevant attribute analysis (ODRA) for local outlier detection in a high-dimensional dataset. There are two phases of the proposed algorithm. During the preliminary stage, we present a data reduction method that reduces the data set by pruning irrelevant attributes and data points. In the second phase, we propose an outlier detection method based on k-NN kernel density estimation. The experimental results on 15 UCI machine learning repository datasets show the supremacy and effectiveness of our proposed approach over state-of-the-art outlier detection methods. [ABSTRACT FROM AUTHOR]

This paper examines the prediction of disturbances based on their past measurements using k -nearest neighbours. The aim is to provide a prediction of a measured disturbance to a controller, in order to improve the feed-forward action. This prediction method works in an unsupervised way, it is robust against changes of the characteristics of the disturbance, and its functioning is simple and transparent. The method is tested on data from industrial process plants and compared with predictions from an autoregressive model. A qualitative as well as a quantitative method for analysing the predictability of the time series is provided. As an example, the method is implemented in an MPC framework to control a simple benchmark model. [ABSTRACT FROM AUTHOR]

Millions of smart meters that are able to collect individual load curves, that is, electricity consumption time series, of residential and business customers at fine scale time grids are now deployed by electricity companies all around the world. It may be complex and costly to transmit and exploit such a large quantity of information, therefore it can be relevant to use survey sampling techniques to estimate mean load curves of specific groups of customers. Data collection, like every mass process, may undergo technical problems at every point of the metering and collection chain resulting in missing values. We consider imputation approaches (linear interpolation, kernel smoothing, nearest neighbours, principal analysis by conditional estimation) that take advantage of the specificities of the data, that is to say the strong relation between the consumption at different instants of time. The performances of these techniques are compared on a real example of Irish electricity load curves under various scenarios of missing data. A general variance approximation of total estimators is also given which encompasses nearest neighbours, kernel smoothers imputation and linear imputation methods. The Canadian Journal of Statistics 47: 65–89; 2019 © 2018 Statistical Society of Canada [ABSTRACT FROM AUTHOR]

Abstract A novel data driven approach to batch process monitoring is presented, which combines the k-Nearest Neighbour rule with the dynamic time warping (DTW) distance. This online method (DTW-NN) calculates the DTW distance between an ongoing batch, and each batch in a reference database of batches produced under normal operating conditions (NOC). The sum of the k smallest DTW distances is monitored. If a fault occurs in the ongoing batch, then this distance increases and an alarm is generated. The monitoring statistic is easy to interpret, being a direct measure of similarity of the ongoing batch to its nearest NOC predecessors and the method makes no distributional assumptions regarding normal operating conditions. DTW-NN is applied to four extensive datasets from simulated batch production of penicillin, and tested on a wide variety of fault types, magnitudes and onset times. Performance of DTW-NN is contrasted with a benchmark multiway PCA approach, and DTW-NN is shown to perform particularly well when there is clustering of batches under NOC. [ABSTRACT FROM AUTHOR]

A biannual survey of physico-chemical quality indices of 104 irrigation-water wells located in a cultivated plain of a Mediterranean island catchment was conducted using a multi-parameter probe. The campaign was planned so as to differentiate between the dry and wet seasons. The acquired data constituted the test bed for evaluating the results and the features of four spatial interpolation methods, i.e. ordinary kriging, universal kriging, inverse distance weighted and nearest neighbours, against those of the recently introduced bilinear surface smoothing (BSS). In several cases, BSS outperformed the other interpolation methods, especially during the two-fold cross-validation procedure. The study emphasizes the fact that both in situ measurements and good mathematical techniques for studying the spatial distribution of water quality indices are pivotal to agricultural practice management. In the specific case studied, the spatio-temporal variability of water quality parameters and the need for monitoring were evident, as low irrigation water quality was encountered throughout the study area. [ABSTRACT FROM AUTHOR]

Fault detection and location are important and front-end tasks in assuring the reliability of power electronic circuits. In essence, both tasks can be considered as the classification problem. This paper presents a fast fault classification method for power electronic circuits by using the support vector machine (SVM) as a classifier and the wavelet transform as a feature extraction technique. Using one-against-rest SVM and one-against-one SVM are two general approaches to fault classification in power electronic circuits. However, these methods have a high computational complexity, therefore in this design we employ a directed acyclic graph (DAG) SVM to implement the fault classification. The DAG SVM is close to the one-against-one SVM regarding its classification performance, but it is much faster. Moreover, in the presented approach, the DAG SVM is improved by introducing the method of K-nearest neighbours to reduce some computations, so that the classification time can be further reduced. A rectifier and an inverter are demonstrated to prove effectiveness of the presented design. [ABSTRACT FROM AUTHOR]

Accurate property valuation is important for property purchasers, investors and for mortgage-providers to assess credit risk in the mortgage market. Automated valuation models (AVM) are being developed to provide cheap, objective valuations that allow dynamic updating of property values over the term of a mortgage. A useful feature of automated valuations is to provide a region of plausible price estimates for each individual property, rather than just a single point estimate. This would allow buyers and sellers to understand uncertainty on pricing individual properties and mortgage providers to include conservatism in their credit risk assessment. In this study, Conformal Predictors (CP) are used to provide such region predictions, whilst strictly controlling for predictive accuracy. We show how an AVM can be constructed using a CP, based on an underlying k-nearest neighbours approach. Time trend in property prices is dealt with by assuming a systematic effect over time and adjusting prices in the training data accordingly. The AVM is tested on a large data set of London property prices. Region predictions are shown to be reliable and the efficiency, ie region width, of property price predictions is investigated. In particular, a regression model is constructed to model the uncertainty in price prediction linked to property characteristics. [ABSTRACT FROM AUTHOR]

• A new approach, GLOBAL-LOCAL, was developed to predict SOC at the local scale from vis–NIR data. • GLOBAL-LOCAL requires a representative set from the samples to be predicted. • GLOBAL-LOCAL improved SOC prediction compared to the entire spectral library. • GLOBAL-LOCAL was more accurate than using site-specific reference samples. • GLOBAL-LOCAL outperformed three other published approaches. There is much interest in the use of large visible near infrared (vis–NIR) soil spectral libraries for rapid soil analysis and to further soils research. However, the use of such large and diverse spectral libraries often provides biased predictions at local scales. Here, we present a new approach, GLOBAL-LOCAL, which aims to use large libraries to develop spectroscopic models that can fit locally. This is achieved by selecting the most common spectral neighbours of a small representative set (Lab) from the target site using a combination of principal component analysis (PCA) and k -nearest neighbours, followed by selection of the calibration with the lowest root mean square error (RMSE) in predicting the Lab , which is then added to the calibration to derive the GLOBAL-LOCAL. We evaluated this approach by predicting the soil organic carbon (SOC) content in two different target sites, one in Canada (Lévis, QC; N = 111) and the other in Finland (Maaninka; N = 101). Partial least squares regression (PLSR) was used for model development and validation. The soil spectral library used was a combination of that developed by the World Agroforestry Center and International Soil Reference Information Centre (N = 3875), and that developed at the Quebec Research and Development Centre (N = 1051). Based on a preliminary analysis, we selected 25 Lab samples from each target site using conditioned Latin hypercube sampling (cLHS). Prediction accuracy by GLOBAL-LOCAL for Lévis and Maaninka (RMSE = 1.41 and 3.08 g kg−1, RPIQ = 5.9 and 4.1, respectively) were better than using the global calibration (RMSE = 20.3 and 23.6 g kg−1, RPIQ = 0.41 and 0.53, respectively). The GLOBAL-LOCAL was better or similar to calibrations with site-specific samples (RMSE = 2.38 and 4.68 g kg−1, RPIQ = 3.5 and 2.7, respectively) and Lab samples only (RMSE = 1.41 and 2.54 g kg−1, RPIQ = 5.9 and 4.9, for Lévis and Maaninka, respectively). In addition, GLOBAL-LOCAL outperformed three other approaches, RS-LOCAL, the spectrum-based learner and LOCAL. While 25 Lab samples was regarded as optimum by cLHS, further examination revealed that as little as 10 Lab samples may be sufficient for the GLOBAL-LOCAL. The method uses the information in the large spectral library so that it can substantially reduce analytical cost and greatly improve predictions on local scales. [ABSTRACT FROM AUTHOR]

Abstract: Human Epithelial (HEp-2) cells are commonly used in the Indirect Immunofluorescence (IIF) tests to detect autoimmune diseases. The diagnosis consists of searching and classification to specific patterns created by Anti-Nuclear Antibodies (ANAs) in the patient serum. Evaluation of the IIF test is mostly done by humans, which means that it is highly dependent on the experience and expertise of the physician. Therefore, a significant amount of research has been focused on the development of computer aided diagnostic systems which could help with the analysis of images from microscopes. This work deals with the design and development of HEp-2 cells classifier. The classifier is able to categorize pre-segmented images of HEp-2 cells into 6 classes. The core of this engine consists of the following image descriptors: Haralick features, Local Binary Patterns, SIFT, surface description and a granulometry-based descriptor. These descriptors produce vectors that form metric spaces. k-NN classification is based on aggregated distance function which combines several features together. An extensive set of evaluations was performed on the publicly available MIVIA HEp-2 images dataset which allows a direct comparison of our approach with other solutions. The results show that our approach is one of the leading classifiers when comparing with other participants in the HEp-2 Cells Classification Contest[1]. [Copyright &y& Elsevier]

Abstract: Nearest neighbour queries (or kNN) have been used in many disciplines, including spatial databases. kNN queries have gone beyond simply finding nearest objects. It is, therefore, important to understand the full spectrum of kNN queries, even before starting to work on how to process and optimise such queries. The aim of this paper is to give a complete picture of what kNN queries are capable of. In this study, we present a survey of kNN queries, in which we propose a taxonomy of kNN queries, comprising four perspectives: (i) a Space perspective, (ii) a Result perspective, (iii) a Query-Point perspective, and (iv) a Relationship perspective. These give a comprehensive overview of kNN queries. As kNN is a large area of research, in this paper, we confine the discussion to kNN queries on stationary objects. [Copyright &y& Elsevier]

In this paper, we consider first an estimator of the residual variance treated by Evans [(2005), ‘Estimating the Variance of Multiplicative Noise’, in 18th International Conference on Noise and Fluctuations, ICNF, in AIP Conference Proceedings, 780, pp. 99–102], Evans and Jones [(2008), ‘Non-Parametric Estimation of Residual Moments and Covariance’, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 464, 2831–2846] and by Liitiäinen, Corona, and Lendasse [(2008), ‘On Nonparametric Residual Variance Estimation’, Neural Processing Letters, 28, 155–167; (2010), ‘Residual Variance Estimation Using a Nearest Neighbour Statistic’, Journal of Multivariate Analysis, 101, 811–823], based on first and second nearest neighbours given an independent and identically distributed sample. Its strong consistency and almost sure convergence of the arithmetic means sequence are shown under mere boundedness and square integrability, respectively, of the response variable Y. Moreover, in view of the local variance, a correspondingly modified estimator of local averaging (partitioning) type is proposed, and strong L 2-consistency for bounded Y, weak L 2-consistency and optimal rate of convergence (for bounded X under suitable Hölder continuity conditions on regression and local variance functions) under moment conditions on Y are established. Simulation studies illustrate the behaviour of the local variance estimates. [ABSTRACT FROM AUTHOR]

K-nearest neighbours algorithms are among the most popular existing classification methods, due to their simplicity and good performances. Over the years, several extensions of the initial method have been proposed. In this paper, we propose a K-nearest neighbours approach that uses the theory of imprecise probabilities, and more specifically lower previsions. We show that the proposed approach has several assets: it can handle uncertain data in a very generic way, and decision rules developed within this theory allow us to deal with conflicting information between neighbours or with the absence of close neighbour to the instance to classify. We show that results of the basic k-NN and weighted k-NN methods can be retrieved by the proposed approach. We end with some experiments on the classical data sets. [ABSTRACT FROM AUTHOR]

Query expansion is the process of reformulating an original query to improve retrieval performance in information retrieval systems. Relevance feedback is one of the most useful query modification techniques in information retrieval systems. In this paper, we introduce query expansion into ligand-based virtual screening (LBVS) using the relevance feedback technique. In this approach, a few high-ranking molecules of unknown activity are filtered from the outputs of a Bayesian inference network based on a single ligand molecule to form a set of ligand molecules. This set of ligand molecules is used to form a new ligand molecule. Simulated virtual screening experiments with the MDL Drug Data Report and maximum unbiased validation data sets show that the use of ligand expansion provides a very simple way of improving the LBVS, especially when the active molecules being sought have a high degree of structural heterogeneity. However, the effectiveness of the ligand expansion is slightly less when structurally-homogeneous sets of actives are being sought. [ABSTRACT FROM AUTHOR]

HSQC spectra are routinely acquired for chemical structure analysis based on hydrogen and carbon chemical environments. Two fast HSQC peak matching algorithms have been developed; a nearest neighbour approach and a probabilistic method based on an existing discrete genetic algorithm. Both of these techniques are intended to find HSQC spectra matches that supplement information generated by established molecular fingerprint methods. Our results are compared to those calculated using a specific implementation of molecular fingerprints. The nearest neighbour and genetic algorithm-based methods ranked highly particular structures missed by molecular fingerprints. Our analysis shows that by complementing molecular fingerprint matches with our findings, a comprehensive list of matches can be identified. The refined list of compounds could be used to improve the quality of compounds used in screening libraries in the pharmaceutical industry. [ABSTRACT FROM AUTHOR]

Abstract: Nearest neighbour (NN) approaches are inspired by the way humans make decisions, comparing a test object to previously encountered samples. In this paper, we propose an NN algorithm that uses the lower and upper approximations from fuzzy-rough set theory in order to classify test objects, or predict their decision value. It is shown experimentally that our method outperforms other NN approaches (classical, fuzzy and fuzzy-rough ones) and that it is competitive with leading classification and prediction methods. Moreover, we show that the robustness of our methods against noise can be enhanced effectively by invoking the approximations of the Vaguely Quantified Rough Set (VQRS) model, which emulates the linguistic quantifiers “some” and “most” from natural language. [Copyright &y& Elsevier]

Cheminformatics protocols have been developed and assessed that identify a small set of fragments which can represent the compounds in a chemical library for use in fragment-based ligand discovery. Six different methods have been implemented and tested on Input Libraries of compounds from three suppliers. The resulting Fragment Sets have been characterised on the basis of computed physico-chemical properties and their similarity to the Input Libraries. A method that iteratively identifies fragments with the maximum number of similar compounds in the Input Library (Nearest Neighbours) produces the most diverse library. This approach could increase the success of experimental ligand discovery projects, by providing fragments that can be progressed rapidly to larger compounds through access to available similar compounds (known as SAR by Catalog). [ABSTRACT FROM AUTHOR]

Abstract: An analytical result should be expressed as x ± U, where x is the experimental result obtained for a given variable and U is its uncertainty. This uncertainty is rarely taken into account in supervised classification. In this paper, we propose to include the information about the uncertainty of the experimental results to compute the reliability of classification. The method combines k-nearest neighbours (kNN) with a nested bootstrap scheme, in which a new bootstrap training set is generated using the classical bootstrap in the first level (B times) and a new bootstrap method, called U-bootstrap, in the second level (D times). Two bootstraps are used to reduce the effect of sampling in the first level and the effect of the uncertainty in the second one. These B × D new training bootstrap sets are used to compute the reliability of classification for an unknown object using kNN. The object is classified into the class with the highest reliability. In this method, unlike the classical kNN and Probabilistic Bagged k-nearest neighbours (PBkNN), the reliability of classification changes (increases or decreases) when the uncertainty is increased. These changes depend on the position of the unknown object with respect to the training objects. For the benchmark Wine dataset, we found similar values of classification error rate (CER) than for kNN (5.57%), but lower than Probabilistic Bagged k-nearest neighbours using Hamamoto''s bootstrap (7.96%) or Efron''s bootstrap (8.97%). [Copyright &y& Elsevier]

Abstract: This paper presents a new method for computing the probability of correct classification for the k-Nearest Neighbours (kNN) method. The method uses bootstrap to provide the posterior probability which a new object is classified with. This is a measure of the reliability of the classification; it increases as the test object is closer to the training objects of a given class and is more sensitive to the position of the test object in the calibration space than the classical measure of posterior probability in kNN. This reliability of the classification is also used to derive a new rule for classification. [Copyright &y& Elsevier]

Abstract: This paper describes an algorithm for the recognition of the type of grip being formed by a subject wearing an intelligent orthotic glove. The signal is a growing multivariate sequence to be correctly classified as soon as the data permit, so the task may be called one of sequential classification or sequential discrimination. The algorithm uses a variant of the ‘ nearest neighbours’ principle with the dissimilarity measure being an exponentially weighted moving average of distances of the Mahalanobis type. The approach taken seems suitable for other problems of this kind. [Copyright &y& Elsevier]

Associated with the establishment or removal of water sources in savanna ecosystems is the issue of the effects of such management actions on animal movement and habitat selection, longer term implications on population levels, and impacts of such change on habitat degradation and soil erosion. Extant metrics used to describe the spatial distribution of water sources on the landscape often fall short of providing source-specific information, making them hard to apply in small-scale management settings. Using the Klaserie Private Nature Reserve (KPNR) as a case study, we compare a buffer framework, describing distances to water, a nearest neighbour framework,and a spatial location-allocation framework (SLAF) created in a geographic information system (GIS). These three frameworks can be combined into one GIS to demonstrate site-specific information on water source distribution, in addition to system-wide descriptions. The visually accessible quality of a GIS allows qualitative input from managers and property owners to achieve quantifiable management goals. The duality of database and visual representation provides a useful tool to assess the role of individual water sources and can easily be updated to reflect changes in their distribution. [ABSTRACT FROM AUTHOR]

1. The composition of freshwater invertebrate assemblages at a location is determined by a range of physico-chemical and biotic factors in the local environment, as well as larger-scale spatial factors such as sources of recruits. We assessed the relative importance of the species composition of local neighbourhoods and proximal environmental factors on the composition of invertebrate assemblages.2. Macroinvertebrate assemblages were sampled at 188 running-water sites in the catchment of the River Rede, north-east England. A total of 176 species were recorded.3. Environmental data, in the form of 13 biotic and abiotic measurements that described stream physical structure, aquatic vegetation and water characteristics, were recorded for each site. Detrended correspondence analysis was then used to simplify nine of these stream environmental variables to create an index of stream structure.4. The species composition of the invertebrate assemblages was related to the environmental variables, using an information theoretic approach. The impact of the species composition of neighbouring sites on each site was determined using Moran's I and autoregressive modelling techniques.5. Species composition was primarily associated with water pH and stream structure. The importance of the species composition of neighbouring sites in determining local species assemblages differed markedly between taxa. The autoregressive component was low for Coleoptera, intermediate for Trichoptera and Plecoptera, and high for Ephemeroptera.6. We hypothesise that the observed differences in the autoregressive component amongst these orders reflects variation in their dispersal abilities from neighbouring sites. [ABSTRACT FROM AUTHOR]

• A new algorithm for performing one-class classification. • First large-scale evaluation of one-class classification without tuning. • Identification of recommended hyperparameter values for range of algorithms. One-class classification is a challenging subfield of machine learning in which so-called data descriptors are used to predict membership of a class based solely on positive examples of that class, and no counter-examples. A number of data descriptors that have been shown to perform well in previous studies of one-class classification, like the Support Vector Machine (SVM), require setting one or more hyperparameters. There has been no systematic attempt to date to determine optimal default values for these hyperparameters, which limits their ease of use, especially in comparison with hyperparameter-free proposals like the Isolation Forest (IF). We address this issue by determining optimal default hyperparameter values across a collection of 246 one-class classification problems derived from 50 different real-world datasets. In addition, we propose a new data descriptor, Average Localised Proximity (ALP) to address certain issues with existing approaches based on nearest neighbour distances. Finally, we evaluate classification performance using a leave-one-dataset-out procedure, and find strong evidence that ALP outperforms IF and a number of other data descriptors, as well as weak evidence that it outperforms SVM, making ALP a good default choice. [ABSTRACT FROM AUTHOR]

This paper deals with the largest interior angle property of the Voronoi nearest neighbours and the Delaunay triangulation in two dimensions. It is indicated that each interior angle opposite to an edge of a Delaunay triangle is the largest of the interior angles opposite the same edge of all possible triangles on the same side. Based on this property, the algorithms for finding the Voronoi neighbours of a point (site) and for constructing the Delaunay triangulation are presented. The algorithms are especially useful for the meshless methods developed in recent years. © 1998 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

• L 1 -CSRT is proposed for one sample per person and based on sparse representation. • Nearest neighbours are identified using kernel-based distance. • It utilizes constrained L 1 -Cuckoo Search Algorithm for coefficient (λ) evaluation. • Optimizing λ increases sparsity and provides higher discrimination ability. • It is analysed with an improvement of 2–6 % in terms of classification accuracy. The performance of face recognition (FR) system essentially depends on the availability of training data. With limited training data available, it creates a major challenge in real world applications. This motivates researchers to investigate for such technique which generates optimum results using limited training samples or even single sample per person (SSPP). In this paper, a novel FR method is developed as Constrained L 1 -Optimal Sparse Representation Technique ( L 1 -CSRT) for SSPP. An optimal sparse representation technique is formulated using the constrained Cuckoo search algorithm (CSA) for estimation of λ coefficients. Further, a novel fitness function is developed based on the L 1 -norm for better classification accuracy yielding better FR. The motivation behind the optimization of λ coefficients using CSA is to increase the sparsity by a better exploration of search space. The efficiency and accuracy of proposed L 1 -CSRT is shown based on the extensive simulations carried out on standard databases. The experimental results are compared with existing methods in terms of mean classification error. The performance of L 1 -CSRT is analysed with an improvement of 2--6 % in terms of classification accuracy. [ABSTRACT FROM AUTHOR]