Search

When recognizing targets by unmanned aerial vehicles (UAVs), problems such as small size, dense dispersion, and complex background are likely to occur, resulting in low recognition rates. In order to solve the above problems, this work proposed a lightweight small target detection algorithm based on the YOLOv8n: STD-YOLOv8 algorithm. First, the regression problem of small targets in different training periods was optimized, the penalty term in the original loss was improved, and a new LIoU loss function was proposed, so that the size of the penalty term could be dynamically adjusted before and after training, thus improving the performance of the algorithm. Second, in order to better adapt to the small target scale and enhance the ability of small target feature acquisition, the SPD-Conv module was integrated in the backbone network, replacing the original stepwise convolutional layer and pooling layer, so as to solve the problems of loss of fine-grained information and low efficiency of feature representation existing in the current convolutional neural network (CNN) structure. In the neck part, nearest-neighbor upsampling was replaced by the feature reassembly assembly of features operator CARAFE (content-aware reassembly of features), which enabled the model to aggregate contextual information in a larger perceptual field and enhanced the feature representation in the neck. Finally, validation experiments were conducted by comparing different algorithms under the same VisDrone-2021 dataset. The results of the ablation experiments show that the algorithms proposed in this thesis have improved the recall (R), mAP50, and mAP95 by 4.7, 5.8 and 5.7%, respectively, compared with YOLOv8n. The results of the model generalization experiments on the TinyPerson dataset show that the algorithm in this paper has superior small target detection performance with only 1.2 M model parameters (1 M = 106). [ABSTRACT FROM AUTHOR]

Estimation of causal effects from observational data has been the main objective in several high-impact scientific domains, while the golden standard for calculating the true causal effect is through the conduction of randomized controlled trials. The abundance of this type of data, which are continuously produced and collected, makes them potentially valuable for estimating causal effects. However, observational data may lead to erroneous treatment effect estimation, since they often suffer from various forms of bias. A recent work has shown that causal effect estimation provided by neural network-based model can be improved by leveraging information from the outcome of neighboring instances in the covariate space. In this work, we propose an information-theoretic methodology for selecting the neighbors to be considered in the estimation of the treatment effect for each sample. The proposed methodology, named Mutual Information-based Neighbor selection for Treatment effect estimation (MINT), selects the optimal number of neighbors as well as the type of distance metric with respect to pre-defined criteria. Then, the average outcome of the neighbors in the treatment and control groups is used as an informative input to the estimator, in addition to the covariates. The presented numerical experiments demonstrate that the adoption of the proposed MINT methodology with the state-of-the-art Dragonnet model is able to develop a reliable and accurate model for treatment effect estimation. [ABSTRACT FROM AUTHOR]

Renewable energy forecasting services comprise various modules for intra-day and day-ahead forecasts. This work specifically addresses day-ahead forecasts, utilizing specifications based on endogenous, historical measurements. These specifications are designed to be computationally efficient, requiring fewer input variables and less training data. Such weather-independent specifications serve as benchmarks against the more computationally demanding forecasts based on numerical weather predictions. A series of experiments, designed to simulate the real-world application of an online system, were conducted on sliding windows of back-contact photovoltaic (installed at KAUST, Saudi Arabia) output series, solar irradiance recorded in Hawaii, and simulated data. Our analysis evaluated 24 specifications, which are variants of (i) functional time series models (including two novel shrinkage procedures); (ii) time series nearest neighbor schemes; (iii) exponential smoothing procedures; (iv) autoregressive integrated moving average processes; (v) automatic techniques based on time series decomposition; and (vi) the persistence model. In addition to employing outlier-robust accuracy metrics, such as mean absolute error, our evaluation also prioritized prediction-interval accuracy, quantified by the mean scaled interval score. Our findings suggest that practitioners can achieve significant improvements over the persistence model by forecasting daily profiles using adaptive nonparametric or functional data analysis-based procedures. Moreover, applying shrinkage to nearest neighbor (NN) forecasts toward smooth, average daily profiles significantly enhances NN performance. Conversely, some popular, computationally intensive models fail to perform adequately to justify their additional cost. [ABSTRACT FROM AUTHOR]

Copyright of Nativa is the property of Revista Nativa and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Indonesia, with 2/3 of its territory being oceans, has the potential to have abundant natural resources in the form of salt. PT. XYZ is the only state-owned company that produces products salt start from process upstream until Downstream, with Wrong One product flagship is salt diet. The biggest challenge in Supply Chain Management at the company is capability compete in providing competitive prices so it is necessary to properly study the costs in the process of distribution and logistics do not charge much in the components of the Cost of Good sold . In study This discuss related determination router best with use method saving Matrix by comparing the Neasrest Insert and Nearest Neighbor algorithms, which is then carried out analysis using Fishbone Diagram in mapping the source of constraints in the distribution process and logistics that focus on 5M+1E elements. The results of this study can be determined by 4 shipping routes according to the Nearest Neighbor algorithm . Improvements to problems that arise in existing conditions can be carried out by adding tools and infrastructure to the loading process facilities, such as hand pallets, forklifts as well as the construction of a loading dock including SOPs that regulate and can be used to measure effectiveness process loading. Indonesia, with 2/3 of its territory being oceans, has the potential to have abundant natural resources in the form of salt. PT. XYZ is the only state-owned company that produces products salt start from process upstream until Downstream, with Wrong One product flagship is salt diet. The biggest challenge in Supply Chain Management at the company is capability compete in providing competitive prices so it is necessary to properly study the costs in the process of distribution and logistics do not charge much in the components of the Cost of Good sold . In study This discuss related determination router best with use method saving Matrix by comparing the Neasrest Insert and Nearest Neighbor algorithms, which is then carried out analysis using Fishbone Diagram in mapping the source of constraints in the distribution process and logistics that focus on 5M+1E elements. The results of this study can be determined by 4 shipping routes according to the Nearest Neighbor algorithm . Improvements to problems that arise in existing conditions can be carried out by adding tools and infrastructure to the loading process facilities, such as hand pallets, forklifts as well as the construction of a loading dock including SOPs that regulate and can be used to measure effectiveness process loading. Keywords : Salt, logistics, diagram fishbone, saving matrix, nearest insert, nearest neighbors [ABSTRACT FROM AUTHOR]

Approximate nearest neighbor search (ANNS) for high dimensional data has received extensive research efforts. Many existing ANNS methods in metric spaces are essentially based on the permutation of pivots, or pre-selected reference points, which are ordered by their distances to the data. Unfortunately, most of these permutation-based methods are quite memory demanding, because of either complexity of index structure, excessive pivots, or insufficient exploitation of distances. This paper proposes the reduced permutation array (RPA), a delicate integration of existing technical elements based on in-depth understanding, for recall@R ANNS in metric spaces. As an array, for each data element, RPA stores only l identities (IDs) of its nearest pivots out of k pre-selected ones (l « k). During search, the order of distances from each data element to the query is approximated by a score function based on the distances from its nearest pivots to the query, and a bounded heap is maintained to store R candidate results. RPA is simple, memory-efficient, and scalable. Experiments show that RPA is approximately 3 times more memory-efficient than existing methods on average. The research results can provide an effective ANNS method for massive data in a single-machine environment with limited memory resources. [ABSTRACT FROM AUTHOR]

It is essential to define more convincing and applicable classifiers for small datasets. In this paper, a minimum reconstruction error-based K-nearest neighbors (K-NN) classifier is proposed. We propose a new neighbor selection method. In the proposed neighbor selection method, a subset of data that minimize the reconstruction error is assigned as the neighbors of the query sample. Also, there is not any constraint on the distance of the neighbors from the query sample. An l 0 -based sparse representation problem is introduced for selecting the neighbors. These neighbors are assigned as neighbors of the minimum reconstruction error-based K-NN classifiers. Three l 0 -based minimum reconstruction error-based K-NN classifiers are introduced. These classifiers are less sensitive to the reconstruction coefficients in minimum reconstruction error-based K-NN classifiers and reconstruct the query sample with less error. The results on UCI machine learning repository, UCR time-series archive datasets, and a small subset (16%) of MNIST handwritten digit database demonstrate the suitable performance of the proposed method. The recognition precision increases up to more than 3% in some evaluations. [ABSTRACT FROM AUTHOR]

We study q-Laguerre multiple orthogonal polynomials. These polynomials are orthogonal with respect to q-analogues of Laguerre weight functions. We focus our attention on their structural properties. Raising and lowering operators as well as Rodrigues-type formulas are obtained and their explicit representations are given. A high-order linear q-difference equation with polynomial coefficients is deduced. Moreover, we obtain the nearest neighbor recurrence relation using a q-analogue of the theorem 23.1.11 by M.E.H. Ismail in [12]. [ABSTRACT FROM AUTHOR]

This paper introduces a novel modification to the Inverted File (IVF) index approach for approximate nearest neighbor search, incorporating supervised learning techniques to enhance the efficacy of intermediate clustering and achieve more balanced cluster sizes. The proposed method involves creating clusters using a neural network by solving a task to classify query vectors into the same bucket as their corresponding nearest neighbor vectors in the original dataset. When combined with minimizing the standard deviation of the bucket sizes, the indexing process becomes more efficient and accurate during the approximate nearest neighbor search. Through empirical evaluation on a test dataset, we demonstrate that the proposed semi-supervised IVF index approach outperforms the industry-standard IVF implementation with fixed parameters, including the total number of clusters and the number of clusters allocated to queries. This novel approach has promising implications for enhancing nearest-neighbor search efficiency in high-dimensional datasets across various applications, including information retrieval, natural language search, recommendation systems, etc. [ABSTRACT FROM AUTHOR]

Collective systems self-organize to form globally ordered spatiotemporal patterns. Finding appropriate measures to characterize the order in these patterns will contribute to our understanding of the principles of self-organization in all collective systems. Here we examine a new measure based on the entropy of the neighbor distance distributions in the characterization of collective patterns. We study three types of systems: a simulated self-propelled boid system, two active colloidal systems, and one centimeter-scale robotic swarm system. In all these systems, the new measure proves sensitive in revealing active phase transitions and in distinguishing steady states. We envision that the entropy by neighbor distance could be useful for characterizing biological swarms such as bird flocks and for designing robotic swarms. [ABSTRACT FROM AUTHOR]

Approximate nearest neighbor (ANN) search has become an essential paradigm for large-scale image retrieval. Conventional ANN search requires the categories of query images to been seen in the training set. However, facing the rapid evolution of newly-emerging concepts on the web, it is too expensive to retrain the model via collecting labeled data with the new (unseen) concepts. Existing zero-shot hashing methods choose the semantic space or intermediate space as the embedding space, which ignore the inconsistency of visual space and semantic space and suffer from the hubness problem on the zero-shot image retrieval task. In this paper, we present an novel deep quantization network with visual-semantic alignment for efficient zero-shot image retrieval. Specifically, we adopt a multi-task architecture that is capable of 1) learning discriminative and polymeric image representations for facilitating the visual-semantic alignment; 2) learning discriminative semantic embeddings for knowledge transfer; and 3) learning compact binary codes for aligning the visual space and the semantic space. We compare the proposed method with several state-of-the-art methods on several benchmark datasets, and the experimental results validate the superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Context. Optimization of the method of nearest neighbors k-NN for the classification of text documents by their topics and experimentally solving the problem based on the method. Objective. The study aims to study the method of nearest neighbors k-NN for classifying text documents by their topics. The task of the study is to classify text documents by their topics based on a dataset for the optimal time and with high accuracy. Method. The k-nearest neighbors (k-NN) method is a metric algorithm for automatic object classification or regression. The k-NN algorithm stores all existing data and categorizes the new point based on the distance between the new point and all points in the training set. For this, a certain distance metric, such as Euclidean distance, is used. In the learning process, k-NN stores all the data from the training set, so it belongs to the “lazy” algorithms since learning takes place at the time of classification. The algorithm makes no assumptions about the distribution of data and it is nonparametric. The task of the k-NN algorithm is to assign a certain category to the test document x based on the categories k of the nearest neighbors from the training dataset. The similarity between the test document x and each of the closest neighbors is scored by the category to which the neighbor belongs. If several of k’s closest neighbors belong to the same category, then the similarity score of that category for the test document x is calculated as the sum of the category scores for each of these closest neighbors. After that, the categories are ranked by score, and the test document is assigned to the category with the highest score. Results. The k-NN method for classifying text documents has been successfully implemented. Experiments have been conducted with various methods that affect the efficiency of k-NN, such as the choice of algorithm and metrics. The results of the experiments showed that the use of certain methods can improve the accuracy of classification and the efficiency of the model. Conclusions. Displaying the results on different metrics and algorithms showed that choosing a particular algorithm and metric can have a significant impact on the accuracy of predictions. The application of the ball tree algorithm, as well as the use of different metrics, such as Manhattan or Euclidean distance, can lead to improved results. Using clustering before applying k-NN has been shown to have a positive effect on results and allows for better grouping of data and reduces the impact of noise or misclassified points, which leads to improved accuracy and class distribution. [ABSTRACT FROM AUTHOR]

The optical properties of Mn4+-activated phosphors are heavily affected by the host and the related data are difficult to obtain, which makes quantitative design of novel phosphors with required wavelength particularly challenging. In the era of big data, computation is emerging as a vital complement to materials development. Compared with traditional calculation routes, machine learning (ML) models provide better solutions for the problems wherein understanding the underlying physics is complicated with sparse data. In this study, we developed six ML models based on a sparse and small data set to rapidly predict the peak emission wavelength of novel phosphors and use Recursive Feature Elimination (RFE) to select the best features. Among them, the K-nearest neighbors regression (KNN.r) algorithm using non-parametric regression technology can predict the value of the target variable based on the similarity between the target value and its spatial neighbors, which can tackle sparse data sets with no obvious linear relationship well and has the best prediction accuracy. These models were then used to predict 278 potential Mn4+-activated phosphor hosts extracted from the inorganic crystal structure database (ICSD), and 19 samples were selected to validate these methods. In order to overcome the lack of physical and chemical insight of KNN, a visual data analysis was carried out based on hierarchical clustering analysis (HCA), and the rule between the peak emission wavelength and the structure of host was proposed. Guided by the results of KNN and HCA, six samples were selected for the synthesis and characterization to verify our conclusions. The measured values of all these phosphors exhibit good agreement with the predicted results, suggesting the success of our methodology. Thus, the model established and the rule proposed in this paper provide a new pathway for the rapid discovery of novel phosphors. [ABSTRACT FROM AUTHOR]

In this paper, we consider a new method dealing with the problem of estimating the scoring function γ a , with a constant a , in functional space and an unknown scale parameter under a nonparametric robust regression model. Based on the k Nearest Neighbors ( k NN) method, the primary objective is to prove the asymptotic normality aspect in the case of a stationary ergodic process of this estimator. We begin by establishing the almost certain convergence of a conditional distribution estimator. Then, we derive the almost certain convergence (with rate) of the conditional median (scale parameter estimator) and the asymptotic normality of the robust regression function, even when the scale parameter is unknown. Finally, the simulation and real-world data results reveal the consistency and superiority of our theoretical analysis in which the performance of the k NN estimator is comparable to that of the well-known kernel estimator, and it outperforms a nonparametric series (spline) estimator when there are irrelevant regressors. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a brain disorder that causes speech and communication problems. The speech issues are often described as slow speech and difficulty with articulation because Jaw muscles don’t move with enough strength. Mainly, clinical experts make a voice assessment to analyze voice signals, to detect PD. In this paper, we designed an intelligent model based on discrete wavelet transform (DWT) for the detection of Parkinson's. Firstly, the voice signals are passed through DWT. The suggested DWT-based model is then used to extract a collection of statistical features from voice data. The extracted features are evaluated, and the significant ones are chosen. The features that have been chosen are fed into a least squares support vector machine (LS-SVM), as well as another four classifiers K-nearest neighbor (KNN), bagged tree, SVM (support vector machine), and k means for the comparison. The proposed model is evaluated using a publicly available dataset named UCI machine learning repository. Several tests were carried out, and the findings revealed that the proposed framework can classify voice signals with a100% accuracy in the LS-SVM technique. [ABSTRACT FROM AUTHOR]

Sudden cardiac arrest (SCA) consisting of ventricular fibrillation and ventricular tachycardia considered as shockable rhythms is a life-threatening heart disease, which is treated efficiently by the automated external defibrillator (AED). This work proposes a novel design of the SAA, which includes a k-nearest neighbors model and a subset of 8 features extracted from the ECG segments, for the SCA diagnosis on the electrocardiogram (ECG) signal. These features are addressed as the most productive subset among 31 input features based on the evaluation of the feature correlation. The recursive feature elimination algorithm combined with the Boosting model and wise-patient fivefold cross-validation method is adopted for the calculation of the average feature importance, which shows the degree of feature correlation, to construct various input feature subsets. Moreover, component feature combinations known as the representatives of the input feature subsets with an enormous level of correlation and independence are transformed from the input subsets by the principal component analysis method. The wise-patient fivefold cross-validation procedure is used for the evaluation of these component feature combinations on the validation set. The proposed SAA is certainly efficient for SCA detection with a small number of the extracted feature and relatively high diagnosis performance such as accuracy of 99.52%, sensitivity of 97.69%, and specificity of 99.91%. [ABSTRACT FROM AUTHOR]

Nearest neighbour similarity measures are widely used in many time series data analysis applications. They compute a measure of similarity between two time series. Most applications require tuning of these measures' meta-parameters in order to achieve good performance. However, most measures have at least O (L 2) complexity, making them computationally expensive and the process of learning their meta-parameters burdensome, requiring days even for datasets containing only a few thousand series. In this paper, we propose UltraFastMPSearch, a family of algorithms to learn the meta-parameters for different types of time series distance measures. These algorithms are significantly faster than the prior state of the art. Our algorithms build upon the state of the art, exploiting the properties of a new efficient exact algorithm which supports early abandoning and pruning for most time series distance measures. We show on 128 datasets from the UCR archive that our new family of algorithms are up to an order of magnitude faster than the previous state of the art. [ABSTRACT FROM AUTHOR]

Predicting compound potency is a major task in computational medicinal chemistry, for which machine learning is often applied. This study systematically predicted compound potency values for 367 target-based compound activity classes from medicinal chemistry using a preferred machine learning approach and simple control methods. The predictions produced unexpectedly similar results for different classes and comparably high accuracy for machine learning and simple control models. Based on these findings, the influence of different data set modifications on relative prediction accuracies was explored, including potency range balancing, removal of nearest neighbors, and analog series-based compound partitioning. The predictions were surprisingly resistant to these modifications, leading to only small error margin increases. These findings also show that conventional benchmark settings are unsuitable for directly comparing potency prediction methods. [ABSTRACT FROM AUTHOR]

Sampling a nearest neighbor is often presented as a Hansen-Hurwitz or Horvitz-Thompson estimation process. This may not be the most informative viewpoint, and measuring the probability of selection is not necessary. The measurement of the nearest object as a "depth" over the selection area can be done by a sampling process, and the total estimated without knowing the polygon areas. The process is unbiased, quite general, and easy to understand. It can be extended to more than just the nearest object to a sample point and to many different polygon shapes. This paper is an extension, simplification and generalization of an earlier paper in this journal (Iles, K. 2009. "Nearest-tree" estimations--A discussion of their geometry, MCFNS 1(2), pp. 47-51.), but does not require a random orientation or weighting for the direction of measurement from the tree to the polygon border. [ABSTRACT FROM AUTHOR]

When missing data are imputed by any method, there is some uncertainty associated with the imputed value. Consequently, when such imputed data are classified, some uncertainty will be propagated to the classifier output. This leads to two issues to address. First, reducing the uncertainty in the imputed value. Second, modeling and processing of the uncertainty associated with the classifier output to arrive at a better decision. To deal with the first issue, we use a latent space representation, while for the second issue we use Dempster-Shafer evidence theory. First, we train a neural network using the data without any missing value to generate a latent space representation of the input. The complete data set is now extended by deleting every feature once. These missing values are estimated using a nearest neighbor-based scheme. The network is then refined using this extended dataset to obtain a better latent space. This mechanism is expected to reduce the effect of the missing data on the latent space representation. Using the latent space representation of the complete data, we train two classifiers, support vector machines and evidential t-nearest neighbors. To classify an input with a missing value, we make a rough estimate of the missing value using the nearest neighbor rule and generate its latent space representation for classification by the classifiers. Using each classifier output, we generate a basic probability assignment (BPA) and all BPAs are combined to get an overall BPA. Final classification is done using Pignistic probabilities computed on the overall BPA. We use three different ways to defining BPAs. To avoid some problems of Dempster's rule of aggregation, we also use several alternative aggregations including some T-norm-based methods. Note that, T-norm has been used for combination of belief function in Pichon and Denœux (in: NAFIPS 2008: 2008 annual meeting of the North American fuzzy information processing society, pp 1–6, 2008). To demonstrate the superiority of the proposed method, we compare its performance with four state-of-the-art techniques using both artificial and real datasets. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) is a platform that manages daily life tasks to establish an interaction between things and humans. One of its applications, the smart office that uses the Internet to monitor electrical appliances and sensor data using an automation system, is presented in this study. Some of the limitations of the existing office automation system are an unfriendly user interface, lack of IoT technology, high cost, or restricted range of wireless transmission. Therefore, this paper presents the design and fabrication of an IoT-based office automation system with a user-friendly smartphone interface. Also, real-time data monitoring is conducted for the predictive maintenance of sensor nodes. This model uses an Arduino Mega 2560 Rev3 microcontroller connected to different appliances and sensors. The data collected from different sensors and appliances are sent to the cloud and accessible to the user on their smartphone despite their location. A sensor fault prediction model based on a machine learning algorithm is proposed in this paper, where the k-nearest neighbors model achieved better performance with 99.63% accuracy, 99.59% F1-score, and 99.67% recall. The performance of both models, i.e., k-nearest neighbors and naive Bayes, was evaluated using different performance metrics such as precision, recall, F1-score, and accuracy. It is a reliable, continuous, and stable automation system that provides safety and convenience to smart office employees and improves their work efficiency while saving resources. [ABSTRACT FROM AUTHOR]

Tri-training expands the training set by adding pseudo-labels to unlabeled data, which effectively improves the generalization ability of the classifier, but it is easy to mislabel unlabeled data into training noise, which damages the learning efficiency of the classifier, and the explicit decision mechanism tends to make the training noise degrade the accuracy of the classification model in the prediction stage. This study proposes the Tri-training algorithm for adaptive nearest neighbor density editing and cross-entropy evaluation (TTADEC), which is used to reduce the training noise formed during the classifier iteration and to solve the problem of inaccurate prediction by explicit decision mechanism. First, the TTADEC algorithm uses the nearest neighbor editing to label high-confidence samples. Then, combined with the relative nearest neighbor to define the local density of samples to screen the pre-training samples, and then dynamically expand the training set by adaptive technique. Finally, the decision process uses cross-entropy to evaluate the completed base classifier of training and assign appropriate weights to it to construct a decision function. The effectiveness of the TTADEC algorithm is verified on the UCI dataset, and the experimental results show that compared with the standard Tri-training algorithm and its improvement algorithm, the TTADEC algorithm has better classification performance and can effectively deal with the semi-supervised classification problem where the training set is insufficient. [ABSTRACT FROM AUTHOR]

Spatio‐temporal datasets that are difficult to analyse are commonly derived from ecological surveys. There are software packages available to analyse these datasets, but many of them require advanced coding skills. There is a growing need for easy‐to‐use packages that researchers can use to analyse common ecological datasets.We develop a particular generalized linear mixed model framework for spatio‐temporal point‐referenced data that is flexible enough to accommodate data from most ecological surveys while being structured enough to facilitate analyses without advanced coding. Our implementation in the starve package uses a computationally efficient version of a nearest‐neighbour Gaussian process enabling analysis of relatively large datasets.A tutorial analysis of a Carolina wren survey presents a recommended workflow for analyses while showcasing the capabilities of the package.Our model and package are tools that can easily be added to researchers' routine to help make sense of data from ecological surveys. We emphasize the ability of our model to create fine‐scale spatio‐temporal predictions which can then be used to visualize and identify important trends in species distributions. [ABSTRACT FROM AUTHOR]

Ground filtering is necessary in processing airborne light detection and ranging (LiDAR) point clouds for forestry applications. This study proposes a progressive plane detection filtering (PPDF) method. First, the method uses multi-scale planes to characterize terrain, i.e., the local terrain with large slope variations is represented by small-scale planes, and vice versa. The planes are detected in local point clouds by the random sample consensus method with decreasing plane sizes. The reliability of the planes to represent local terrain is evaluated and the planes with optimal sizes are selected according to evaluation results. Then, ground seeds are identified by selecting the interior points of the planes. Finally, ground points are iteratively extracted based on the reference terrain, which is constructed using evenly distributed neighbor ground points. These neighbor points are identified by selecting the nearest neighbor points of multiple subspaces, which are divided from the local space with an unclassified point as center point. PPDF was tested in six sites with various terrain and vegetation characteristics. Results showed that PPDF was more accurate and robust compared to the classic filtering methods including maximum slope, progressive morphology, cloth simulation, and progressive triangulated irregular network densification filtering methods, with the smallest average total error and standard deviation of 3.42% and 2.45% across all sites. Moreover, the sensitivity of PPDF to parameters was low and these parameters can be set as fixed values. Therefore, PPDF is effective and easy-to-use for filtering airborne LiDAR data. [ABSTRACT FROM AUTHOR]

Classification of the patterns is a crucial structure of research and applications. Using fuzzy set theory, classifying the patterns has become of great interest because of its ability to understand the parameters. One of the problems observed in the fuzzification of an unknown pattern is that importance is given only to the known patterns but not to their features. In contrast, features of the patterns play an essential role when their respective patterns overlap. In this paper, an optimal fuzzy nearest neighbor model has been introduced in which a fuzzification process has been carried out for the unknown pattern using k nearest neighbor. With the help of the fuzzification process, the membership matrix has been formed. In this membership matrix, fuzzification has been carried out of the features of the unknown pattern. Classification results are verified on a completely llabelled Telugu vowel data set, and the accuracy is compared with the different models and the fuzzy k nearest neighbor algorithm. The proposed model gives 84.86% accuracy on 50% training data set and 89.35% accuracy on 80% training data set. The proposed classifier learns well enough with a small amount of training data, resulting in an efficient and faster approach. [ABSTRACT FROM AUTHOR]

In this work, we study the effects of geometric confinement on the point statistics in a quasi-low-dimensional system. Specifically, we focus on the nearest-neighbor statistics. Accordingly, we have performed comprehensive numerical simulations of binomial point process on quasi-one-dimensional rectangle strips for different values of the confinement ratio defined as the ratio of the strip width to the mean nearest-neighbor distance. We found that the nearest-neighbor distance distributions (NNDDs) conform to an extreme value Weibull distribution with the shape parameter depending on the confinement ratio, while the process intensity remains constant. This finding reveals the reduction of effective spatial degrees of freedom in a quasi-low-dimensional system under the geometric confinement. The scale dependence of the number of effective spatial degrees of freedom is found to obey the crossover ansatz. We stress that the functional form of the crossover ansatz is determined by the nature of the studied point process. Accordingly, different physical processes in the quasi-low-dimensional system obey different crossover ansatzes. The relevance of these results for quasi-low-dimensional systems is briefly highlighted. [ABSTRACT FROM AUTHOR]

Reversed-Phase Liquid Chromatography (RPLC) is a common liquid chromatographic mode used for the control of pharmaceutical compounds during their drug life cycle. Nevertheless, determining the optimal chromatographic conditions that enable this separation is time consuming and requires a lot of lab work. Quantitative Structure Retention Relationship models (QSRR) are helpful for doing this job with minimal time and cost expenditures by predicting retention times of known compounds without performing experiments. In the current work, several QSRR models were built and compared for their adequacy in predicting the retention times. The regression models were based on a combination of linear and non-linear algorithms such as Multiple Linear Regression, Support Vector Regression, Least Absolute Shrinkage and Selection Operator, Random Forest, and Gradient Boosted Regression. Models were built for five pH conditions, i.e., at pH 2.7, 3.5, 6.5, and 8.0. In the end, the model predictions were combined using stacking and the performances of all models were compared. The k-nearest neighbor-based application domain filter was established to assess the reliability of the prediction for further compound prioritization. Altogether, this study can be insightful for analytical chemists working with RPLC to begin with the computational prediction modeling such as QSRR to predict the separation of small molecules. [ABSTRACT FROM AUTHOR]

In the previous research on traffic flow prediction models, most of the models mainly studied the time series of traffic flow, and the spatial correlation of traffic flow was not fully considered. To solve this problem, this paper proposes a method to predict the spatio-temporal characteristics of short-term traffic flow by combining the k-nearest neighbor algorithm and bidirectional long short-term memory network model. By selecting the real-time traffic flow data observed on high-speed roads in the United Kingdom, the K-nearest neighbor algorithm is used to spatially screen the station data to determine the points with high correlation and then input the BILSTM model for prediction. The experimental results show that compared with SVR, LSTM, GRU, KNN-LSTM, and CNN-LSTM models, the model proposed in this paper has better prediction accuracy, and its performance has been improved by 77%, 19%, 18%, 22%, and 13%, respectively. The proposed K-nearest neighbor-bidirectional long short-time memory model shows better prediction performance. [ABSTRACT FROM AUTHOR]

In recent years, collaborative representation (CR)-based models have been widely used in hyperspectral image (HSI) classification. And many CR models utilize Tikhonov regularization term to improve the classification performance. However, this regularization term greatly increases the running time of CR models due to calculating the Euclidean distance between the test sample and all the training samples. Moreover, most of CR models utilize the training samples of all classes to represent the test sample, which may make the classes irrelevant to the test sample produce an impact on the classification performance of CR. To solve the above problems and further improve the classification performance of the CR model, a novel Euclidean distance-based adaptive (EDA) dictionary learning method is proposed in this paper, which aims to construct a subdictionary by selecting the k-nearest neighbor classes relevant to the test sample using Euclidean distance. And the EDA method is integrated into the original collaborative representation classifier (CRC) and the CRC with Tikhonov regularization (CRT) methods for HSI classification, which is denoted as EDACRC and EDACRT, respectively. The proposed methods are evaluated on the Botswana, KSC, and Salinas-A datasets, in which the classification performances and running times of these methods are analyzed and compared. The experimental results demonstrate that the proposed EDA method can not only effectively reduce the running time of the CR model with Tikhonov regularization, but also effectively eliminate some classes irrelevant to the test sample, thus improving the classification performance of the CR model. [ABSTRACT FROM AUTHOR]

Ensemble diversity is an important characteristic of Multiple Classifier Systems (MCS), which aim at improving the overall performance of a classification system by combining the response of several models. While diversity may be introduced through various manipulations at the data level and the model level, some MCSs incorporate local information in order to increase it and/or take advantage of it, based on the idea that the different classifiers in the ensemble may have expertise in distinct areas of the feature space. 1 1 Code available at: https://github.com/marianaasouza/olp%5fplusplus. Following a similar reasoning, we introduced in a previous work an ensemble method which produces in test time a few experts in the local region where each given query sample is located. These local experts, which are generated with slightly differing views of the target area, are then used to label the corresponding unknown instance. While the framework was shown to perform well especially over imbalanced problems, the locality definition in the method is based on the nearest neighbors rule and Euclidian distance, as is the case of various local-based ensembles, which may suffer from the effects of the curse of dimensionality over high dimensional problems. Thus, in this work, we propose a local ensemble method in which we leverage the data partitions given by decision trees for locality definition. More specifically, the partitions defined at different levels of the decision path that a given query instance traverses in the tree(s) are used as the regions over which the local experts are produced. By using different node levels from the path, each classifier in the local pool has a moderately distinct view of the target region without resorting to a dissimilarity metric, which might be susceptible to high dimensional spaces. Experimental results over 39 high dimensional problems showed that the proposed approach was significantly superior to our previous, distance-based framework in balanced accuracy rate. Compared to other six local-based ensemble methods, including dynamic selection and weighting schemes, the proposed method achieved competitive results, outperforming the random forest baseline and two state-of-the-art dynamic ensemble selection techniques. • Local-based ensemble methods produce and/or exploit models specialized in each area. • Locality is often achieved via distance metrics which degrade in high dimensionality. • We propose a method that produces local experts on the fly for high dimensional data. • The models are fit over multiple partitions recursively defined by decision trees. • The proposed method outperformed several distance-based local ensemble methods. [ABSTRACT FROM AUTHOR]

Data-driven preventive scanning for transient stability assessment (DTSA) is a faster and more efficient solution than time-domain simulation (TDS). However, most current methods cannot balance generalization to different topologies and interpretability, with simple output. A model that conforms to the physical mechanism and richer label for transient stability can increase confidence in DTSA. Thus a static-information, k-neighbor, and self-attention aggregated schema (SKETCH) is proposed in this paper. Taking only static measurements as input, SKETCH gives several explanations that are consistent with the physical mechanisms of TSA and provides results for all generator stability while predicting system stability. A module based on the self-attention mechanism is designed to solve the locality problem of a graph neural network (GNN), achieving subgraph equivalence outside the k-order neighborhood. Test results on the IEEE 39-bus system and IEEE 300-bus system indicate the superiority of SKETCH and also demonstrate the rich sample interpretation results. Highlights: A fast TSA scheme for pre-failure scanning. A physical mechanism-based attention structure for dynamic graph pooling. A node regression model that responds to key physical mechanisms. Generator label for richer output information. Top performance and post-hoc interpretation. [ABSTRACT FROM AUTHOR]

With the wide spread of massive open online courses (MOOC), millions of people have enrolled in many courses, but the dropout rate of most courses is more than 90%. Accurately predicting the dropout rate of MOOC is of great significance to prevent learners' dropout behavior and reduce the dropout rate of students. Using the PH278x curriculum data on the Harvard X platform in spring 2013, and based on the statistical analysis of the factors that may affect learners' final completion of the curriculum from two aspects: learners' own characteristics and learners' learning behavior, we established the MOOC dropout rate prediction models based on logical regression, K nearest neighbor and random forest, respectively. Experiments with five evaluation metrics (accuracy, precision, recall, F1 and AUC) show that the prediction model based on random forest has the highest accuracy, precision, F1 and AUC, which are 91.726%, 93.0923%, 95.4145%, 0.925341, respectively, its performance is better than that of the prediction model based on logical regression and that of the model based on K-nearest neighbor, whose values of these metrics are 91.395%, 92.8674%, 95.2337%, 0.912316 and 91.726%, 93.0923%, 95.4145% and 0.925341, respectively. As for recall metrics, the value of random forest is higher than that of KNN, but slightly lower than that of logistic regression, which are 0.992476, 0.977239 and 0.978555, respectively. Then, we conclude that random forests perform best in predicting the dropout rate of MOOC learners. This study can help education staff to know the trend of learners' dropout behavior in advance, so as to put some measures to reduce the dropout rate before it occurs, thus improving the completion rate of the curriculum. [ABSTRACT FROM AUTHOR]

This article proposes an open-set recognition model that is based on the use of extreme value statistics. For this purpose, a distance ratio is introduced that expresses how dissimilar a target point is from the known classes by considering the ratio of distances locally around the target point. It is shown that the class of generalized Pareto distributions with bounded support can be used to model the peaks of the distance ratio above a high threshold. The resulting distribution provides a probabilistic framework to perform open-set recognition. Furthermore, we describe a numerical procedure to estimate the hyperparameters of our model. This procedure is based on a new objective function that considers both the fit of the generalized Pareto distribution and the misclassification error of the known classes. Our method is applied to three image datasets and an audio dataset showing that it outperforms similar open-set recognition and anomaly detection methods. for this article are available online. [ABSTRACT FROM AUTHOR]

Studies on safety in aviation are necessary for the development of new technologies to forecast and prevent aeronautical accidents and incidents. When predicting these occurrences, the literature frequently considers the internal characteristics of aeronautical operations, such as aircraft telemetry and flight procedures, or external characteristics, such as meteorological conditions, with only few relationships being identified between the two. In this study, data from 6,188 aeronautical occurrences involving accidents, incidents, and serious incidents, in Brazil between January 2010 and October 2021, as well as meteorological data from two automatic weather stations, totaling more than 2.8 million observations, were investigated using machine learning tools. For data analysis, decision tree, extra trees, Gaussian naive Bayes, gradient boosting, and k-nearest neighbor classifiers with a high identification accuracy of 96.20% were used. Consequently, the developed algorithm can predict occurrences as functions of operational and meteorological patterns. Variables such as maximum take-off weight, aircraft registration and model, and wind direction are among the main forecasters of aeronautical accidents or incidents. This study provides insight into the development of new technologies and measures to prevent such occurrences. [ABSTRACT FROM AUTHOR]

Children with autism spectrum disorders (ASDs) tremendously impact people's lives, and the incidence and prevalence of ASDs are increasing globally. Global health organisations and other autism-treatment centres specialising in autism diagnosis and detection are encountering challenges on how to provide an appropriate ASD diagnosis system that enables accurate analyses and early detection of autism. Information about ASD detection is affected by unknown aetiology of the disease, and an urgent solution is required to investigate its aetiological factors. Accordingly, increasing the opportunities to provide evidence of the 'sociodemographic and family characteristics' risk factors in predicting ASD is a scientific complex problem that needs to be solved. This study developed an early prediction model for diagnosing and detecting children with ASD based on effective sociodemographic and family characteristic features related to ASD using the machine learning (ML) model. The proposed methodology involves three phases. The identification phase is first accomplished by identifying a large-scale ASD dataset and preprocessing stages: 1-NN model for imputing missing data, feature-selection methods using Chi2 and Relief, and adaptive balancing data approach using Synthetic Minority Oversampling Technique. Chi2 and Relief are applied to determine the most effective sociodemographic and family characteristic features and produce a new balanced ASD dataset. The second development phase trains and tests the newly prepared ASD dataset through eight ML methods: decision tree, random forest, Naive Bayes, kNN, SVM, logistic regression, AdaBoost, and neural network multilayer perceptron (MLP). The developed model is evaluated in the third phase using five metrics: accuracy, precision, recall, F1 and AUROC, and test time in seconds. Results indicated the following: (1) Out of 10 highly effective sociodemographic and family characteristic features, seven related to autism cases are extracted. (2) Correlation sensitivity analysis reveals that the 'Mom_age_at_child_birth' has the highest positive correlation with 'Father_age_at_child_birth,' with an r-value of 0.751. Moreover, 'child_birth_month' and 'Birth_number' have the highest negative correlation with 'Ses_points_1_10', with an r-value of (− 0.07). (3) AdaBoost, neural network, K-nearest neighbour, and decision tree methods show higher accuracy results (0.9995, 0.9925, 0.9834, and 0.9786, respectively), whereas random forest, logistic regression, and Naive Bayes methods show relatively lower accuracy (0.8297, 0.8199 and 0.8002, respectively). However, the support vector machine method shows the lowest accuracy (0.7105). AdaBoost obtained the highest accuracy on the basis of four other evaluation metrics (AUC = 0.9999, F1 = 0.9995, precision = 0.9995 and recall = 0.9995). Accordingly, the new preprocessed and balanced ASD dataset can be utilised as a data source for autism research. The preprocessing stages can be considered correct and successfully perform better results than the original ASD dataset. Similar results from Chi2 and Relief in the feature-selection approaches substantially improved the classification accuracy. The study confirms the efficacy of the proposed prediction model compared with previous models in different comparative points. Early prediction of autism is possible through this proposed model. [ABSTRACT FROM AUTHOR]

In this study, an algorithm based on the Insertion Heuristic is proposed to solve the Travelling Salesman Problem. In the proposed algorithm, firstly the center and center triangle are determined, then vertices are added to the initial tour by starting the furthest vertex from the center. Several examples of traveling salesman problem are solved using the proposed algorithm and results are compared with the Nearest Neighbor and Greedy Algorithms. Computational Experiments show that the proposed algorithm is efficient. [ABSTRACT FROM AUTHOR]

Magnetorheological dampers are highly nonlinear damping elements which exhibit multi-dimensional data correspondence problems in the current input and damping force output data sets. Therefore, it is inconvenient to use the traditional modeling method to predict the output damping force of the magnetorheological damper accurately. Although the neural network can predict data-driven output results to a high precision, more models, parameters, and complex structures are required to predict the expected current from the magnetorheological test data set. In this study, a prediction model of the expected current in a magnetorheological damping system was proposed by using the k-nearest neighbor algorithm based on the classification algorithm in machine learning. Then, the k-nearest neighbor prediction model was trained and tested with data obtained experimentally. In order to verify the performance of the k-nearest neighbor algorithm, a comparison was made with a prediction model based on the BP neural network, and the damping force output of the predicted current was simulated. The study results show that the k-nearest neighbor prediction model established has a more straightforward working principle, fewer input variables, and a higher approximation accuracy compared with the BP neural network control system. [ABSTRACT FROM AUTHOR]

Developing methods of reconstructing historical geographic scenarios is a significant research topic in the field of geographic information science. To reconstruct an archaeological geographic scenario, we adopted 3D laser scanning technology to acquire hierarchical excavation data in accordance with the field archaeology criterion. This technology originated from the laser sensor and it can perform accurate, fast, and noncontact data acquisition in the field of archaeology. The processing of the scanning data is closely related to the accuracy and efficiency of the reconstruction. Our research focused on the methods of multitemporal point cloud data registration, object-oriented target segmentation, and relic feature extraction based on the nearest neighbor search method. In this study, archaeological excavation data acquired in 2015 at the Lingjiatan site in Hanshan Country, Anhui Province, was taken as the research object. The experiment revealed that the proposed methods can realize the efficient and automatic data collection and geometric feature extraction of relics with high feasibility and reliability. The proposed methods are expected to increase the application of multitemporal point cloud data processing and provide basic modeling methods and data for reconstructing historical geographic scenarios. [ABSTRACT FROM AUTHOR]

To address the problem of low detection accuracy due to training noise caused by mislabeling when Tri-training for network intrusion detection (NID), we propose a Tri-training algorithm based on cross entropy and K-nearest neighbors (TCK) for network intrusion detection. The proposed algorithm uses cross-entropy to replace the classification error rate to better identify the difference between the practical and predicted distributions of the model and reduce the prediction bias of mislabeled data to unlabeled data; K-nearest neighbors are used to remove the mislabeled data and reduce the number of mislabeled data. In order to verify the effectiveness of the algorithm proposed in this paper, experiments were conducted on 12 UCI datasets and NSL-KDD network intrusion datasets, and four indexes including accuracy, recall, F-measure and precision were used for comparison. The experimental results revealed that the TCK has superior performance than the conventional Tri-training algorithms and the Tritraining algorithms using only cross-entropy or K-nearest neighbor strategy. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In recent years, there exist few difficulties for textile industries to predict the yarn nep index for small data and data with mutation. To fill this gap, a yarn nep prediction method combining grey correlation analysis and nearest-neighbour prediction method is proposed. In this paper, 26 indicators such as the raw cotton quality indicators and key process parameters are used as the input of the prediction model for yarn nep. The experimental results show that the relative error of the new method is lower than 10%, while the relative error of the individual data predicted by the traditional three-layer BP neural network is very large. Compared with the BP neural network, the average relative error and root-mean-square error of our proposed method are smaller, while the data are stable and the volatility is small. The prediction performance meets the user's requirements. The effectiveness of the proposed model is proved. [ABSTRACT FROM AUTHOR]

Internet of things (IoT) comprises many heterogeneous nodes that operate together to accomplish a human friendly or a business task to ease the life. Generally, IoT nodes are connected in wireless media and thus they are prone to jamming attacks. In the present scenario jamming detection (JD) by using machine learning (ML) algorithms grasp the attention of the researchers due to its virtuous outcome. In this research, jamming detection is modelled as a classification problem which uses several features. Using one/two or minimumnumber of features produces vague results that cannot be explained. Also the relationship between the feature and the class label cannot be efficiently determined, specifically, if the chosen number of features for training is minimum (say 1 or 2). To obtain good results, machine-learning algorithms are trained by large number of data sets. However, collection of large number of datasets to solve jamming detection is not easy and most of the times generation and collection of large data sets become paradigmatic. In this paper, to solve this problem, more number of features with nominal number of data's is considered that eases the data collection and the classification accuracy. In this research, an efficient technique based on locality sensitive hashing (LSH) for K-nearest neighbor algorithm (K-NN), which takes less time for constructing and querying the hash table that gives good accuracy is proposed and evaluated. From the results, it is clear that the obtained results are validatable and the model is more sensible. [ABSTRACT FROM AUTHOR]

The nearest neighbor model has been a popular model for spatial data. This model assumes a distance‐based neighborhood structure among a set of entities that define the observations in a dataset. Values of a dependent variable are a function of the corresponding values of its neighbors and values of its own explanatory variables. In this paper, we extend the concept of "spatial neighbors" to a more general network dependency where the nearest neighbor is not necessarily distance‐based. As each entity can belong to multiple groups, observations can be related by multiple networks. By exploring the overlapping information that are exhibited by an entity and its neighbors, we modify the traditional spatial autoregressive model with autoregressive disturbances (SARAR) in three ways. First, multiple networks are allowed. Second, differences between the explanatory variables and their corresponding neighboring averages are applied. Third, apart from independent innovations, we investigate the case where the innovations are also multiple‐network dependent. Statistical inference of the model parameters is achieved via maximizing the profile log‐likelihood function. The standard errors of the profile maximum likelihood estimators are calculated by the inverse Fisher information matrix. Four variations of the multiple network autoregressive models are applied to predict changes in stock prices of 555 companies listed in the Hong Kong Stock Exchange. The model with attribute differencing and dependent innovations produces a smaller out‐of‐sample mean prediction error sum of squares than that of the other models. [ABSTRACT FROM AUTHOR]