Nearest-neighbor Classifier Research Articles

Reconstructing Nerve Structures from Unorganized Points

Realistic sensory feedback is paramount for amputees as it improves prosthetic limb control and boosts functionality, safety, and overall quality of life. This sensory restoration relies on the direct electrostimulation of residual peripheral nerves. Computational models are instrumental in simulating these neurostimulation effects, offering solutions to the complexities tied to extensive animal/human trials and costly materials. Central to these models is the detailed mapping of nerve geometry, necessitating the delineation of internal nerve structures, such as fascicles, across various cross-sections. In our modeling process, we faced the challenge of organizing an originally unstructured set of points into coherent contours. We introduced a parameter-free curve-reconstruction algorithm that combines valley-seeking clustering, an adaptive Kalman filter, and the nearest neighbor classification technique. While intuitively simple for humans, the task of reconstructing multiple open and/or closed lines with pronounced corners from a nonuniform point set is daunting for many algorithms. Additionally, the precise differentiation of adjacent curves, commonly encountered in realistic nerve models, remains a formidable challenge even for top-tier algorithms. Our proposed method adeptly navigates the complexities inherent to nerve structure reconstruction. While our algorithm is chiefly designed for closed curves, as dictated by nerve geometry, we believe it can be reconfigured with appropriate code adjustments to handle open curves. Beyond neuroprosthetics, our proposed model has the potential to be applied and spark innovations in biomedicine and a variety of other fields.

Pseudo-label estimation via unsupervised Identity Link Prediction for one-shot person Re-Identification

In this paper, we propose an unsupervised identity link prediction (ILP) method for label estimation in one-shot person Re-ID. ILP aims to relax the constraints of labeled samples and group a set of unlabeled pedestrians by their potential identities. The main idea is that the category relationships between pedestrians (nodes) can be inferred from their local context in the feature space. Specifically, an Identity Link Subgraph (ILS) describes the link relationship between nodes and their nearest neighbors, which is constructed by a two-step procedure. Meanwhile, a Dynamic Penalty Module (DPM) is introduced at each ILS construction step to infer which linkage between pairs in the ILS should be pruned to assign higher-quality classification labels. To fully use the accurate identity information in initial labeled samples, we jointly use identity pseudo-labels (which are estimated by adopting the Nearest Neighbors classifier) with classification pseudo-labels for model training. Moreover, we design a Dual-Branch Fusion network (DBF-Net) to optimize the CNN model simultaneously through all (pseudo-)labeled samples. Results on multiple datasets prove that DBF-Net outperforms traditional one-shot Re-ID methods by a large margin.

Handwritten Devanagari Text Recognition using Single Classifier Approach with VSPCA Scheme

In this research paper we used individual classifier approach for Handwritten Devanagari text recognition. We experimented different categorical classifiers namely Random Forest Classifier (RFC), Support Vector Machine (SVM), K Nearest Neighbor Classifier (KNN), Logistic Regression Classifier (LogRegr), Decision Tree Classifier (DTree). Seven different feature sets are used namely Eccentricity, Euler Number, Horizontal Histogram, Vertical Histogram, HOG Features, LBP Features, and Statistical Features. The experimentation is carried out on 9434 different characters whose features are extracted from 220 handwritten image documents from PHDIndic_11 dataset. We deduced and implemented a unique scheme namely VSPCA scheme. VSPCA is Vectorization, Scaling, and Principal Component Analysis carried out on all feature sets before being given for model training. We obtained varied accuracies using all these five classifiers on all these six feature sets in which 99.52% highest accuracy is observed.

SMOTE-kTLNN: A hybrid re-sampling method based on SMOTE and a two-layer nearest neighbor classifier

In recent years, class-imbalanced learning has become an important branch of machine learning. Synthetic Minority Oversampling Technique (SMOTE) is known as a benchmark method to address imbalanced learning. Although SMOTE performs well on many data, it also has the drawback of generating noisy samples. There are many SMOTE variants to solve this problem. Specifically, these methods are hybrid sampling methods, that is, carrying out an undersampling stage after SMOTE to remove noisy samples. It requires a method that can accurately identify noise to provide reliable performance. In this paper, a hybrid re-sampling method based on SMOTE and a two-layer nearest neighbor classifier (SMOTE-kTLNN) is proposed. SMOTE-kTLNN recognition noise is realized by an Iterative-Partitioning Filter (IPF). Specifically, SMOTE is performed on the original data to balance the data, then the data is divided into n equal parts, establishing kTLNN on each part to predict the whole data. And noisy samples are removed according to the majority voting rule. In the last, the balanced data sets are used to train kNN, AdaBoost, and SVM to verify whether SMOTE-kTLNN is irrelevant to the classifier. The experiment results demonstrate that SMOTE-kTLNN performs better than the comparisons in 25 binary data sets, including Recall, AUC, F1-measure, and G-mean.

Enhancing Image Clarity: A Non-Local Self-Similarity Prior Approach for a Robust Dehazing Algorithm

When light propagates in foggy weather, it is affected and scattered by suspended particles in the air. As a result, images taken in this environment often suffer from blurring, reduced contrast, loss of details, and other issues. The primary challenge in dehazing images is to estimate the transmission coefficient map in the atmospheric degradation model. In this paper, we propose a dehazing algorithm based on the optimization of the “haze-line” prior and non-local self-similarity prior. First, we divided the input haze image into small blocks and used the nearest neighbor classification algorithm to cluster the small patches, which were referred to as “patch-lines”. Based on the characteristics of these “patch-lines”, we could estimate the transmission coefficient map for the image. We then applied the transmission map to a weighted least squares filter to smooth it. Finally, we calculated the clear image using the haze degradation model. The experimental results demonstrate that our algorithm enhanced the image contrast and preserved the fine details, both qualitatively and quantitatively.

Classification of Acoustic Emission Signal Sources Using Genetic Programming

Genetic Programming (GP) is used to classify the Acoustic Emission (AE) signal to different source of signals. GP has the ability to discover the relationship among the observed data and express them mathematically. GP based technique has an advantage over conventional statistical technique like maximum likelihood estimate, nearest neighbor classifier and etc, because they are distribution free i.e. no a priori knowledge is required about the distribution of data. The data is obtained through an AE test using pulse, pencil and spark signal source on the surface of solid steel block. The signal parameters are measured using AET 5000 system, A set of expert system like rules is extracted from the GP. These rules are used directly to classify the AE signals. Classification of AE signal using genetic programming is found to be effective.

Can Artificial Intelligence help a clinical laboratory to draw useful information from limited data sets? Application to mixed connective tissue disease

<p>Diagnosis is a key step of patient management. During decades, refined decision algorithms and numerical scores based on conventional statistical methods were elaborated to ensure optimal reliability. Recently, a number of machine learning tools were developed and applied to process more and more extensive data sets, including up to millions of items and yielding sophisticated classification models. While this approach met with impressive efficiency in some cases, practical limitations stem from the high number of parameters that may be required by a model, resulting in increased cost and delay of decision making. Also, information relative to the specificity of local recruitment may be lost, hampering any simplification of universal models. Here, we explored the capacity of currently available artificial intelligence tools to classify patients found in a single health center on the basis of a limited number of parameters. As a model, the discrimination between systemic lupus erythematosus (SLE) and mixed connective tissue disease (MCTD) on the basis of thirteen biological parameters was studied with eight widely used classifiers (including logistic regression, support vector machine, nearest neighbor classifier, random forests and neural networks). A retrospective study including 44 patients (34 SLE, 10 MCTD) was conducted in Marseilles hospital organization. The best area under ROC curve yielded on test sets with classifiers using all 13 parameters was 0.83 ± 0.03 standard error and 0.86 ± 0.02 SE with 5 selected parameters. It is concluded that classification efficiency may be significantly improved by a knowledge-based selection of discriminating parameters.</p>

Optimal dwelling time prediction for package tour using K‐nearest neighbor classification algorithm

AbstractWe introduce a machine learning‐based web application to help travel agents plan a package tour schedule. K‐nearest neighbor (KNN) classification predicts the optimal tourists' dwelling time based on a variety of information to automatically generate a convenient tour schedule. A database collected in collaboration with an established travel agency is fed into the KNN algorithm implemented in the Python language, and the predicted dwelling times are sent to the web application via a RESTful application programming interface provided by the Flask framework. The web application displays a page in which the agents can configure the initial data and predict the optimal dwelling time and automatically update the tour schedule. After conducting a performance evaluation by simulating a scenario on a computer running the Windows operating system, the average response time was 1.762 s, and the prediction consistency was 100% over 100 iterations.

Rotating Rectifier Fault Diagnosis of Nuclear Multiphase Brushless Excitation System Based on DTW Metric and kNN Classifier

Multi-phase annular brushless excitation system is widely used in large-capacity nuclear power units. Accurate fault diagnosis of the rotating rectifier is of great significance to improve the reliability of the excitation system. However, the traditional diagnosis method based on harmonic analysis of stator field current has some deficiencies. In this paper, a novel diagnosis method using field current waveforms and artificial intelligence is presented. Firstly, the various shape features in field current waveforms of the different rotating rectifier faults are analyzed. Then, the field current waveforms are used as the input of a hybrid algorithm based on the dynamic time warping (DTW) metric and the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> -Nearest Neighbors ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> NN) classifier (DTW- <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> NN). That is, a DTW metric is used to calculate the distance among the shape features in field current waveforms and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> NN classifier is used to diagnose the specific rotating rectifier fault. Finally, experiments on an 11-phase prototype prove the effectiveness of the hybrid method DTW- <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> NN. It is worth mentioning that an improved training set including all trends of field current waveforms should be selected to avoid the asymmetry between each pair of field poles. The learning method provides a new idea for fault diagnosis of the rotating rectifier.

Research on the identification of land types and tree species in the Engebei ecological demonstration area based on GF-1 remote sensing

Identifying forest types is crucial in satellite remote sensing monitoring. Research focusing on the identification of forest tree species using high-resolution remote sensing images is on the rise. Traditional pixel-based classification methods have not been successful in fully utilizing the rich spatial and texture feature information present in the image. Further, these methods are plagued by “classification noise”, which impedes the precise extraction of information regarding forest tree species. The object-oriented classification method, on the other hand, offers an extraction methodology rooted in image segmentation object features. It efficiently leverages the spectral, texture, and spatial geometry information of high-resolution remote sensing images. In this study, an initial application of the object-oriented, multi-level information extraction was conducted on ground objects within the Engebei ecological demonstration area. This was performed utilizing Gaofen (GF)-1 remote sensing images. Following this, a multi-scale segmentation algorithm was utilized to establish different segmentation scales relative to the features of the different objects. The optimal segmentation scale, shape factor, and compactness factor were determined to be 222/0.5/0.5, 167/0.5/0.6, and 81/0.4/0.6, respectively, for the three levels. Subsequently, training samples were used to select optimal features during the classification process. The nearest neighbor classification and random forest classification were performed respectively. The results indicate an overall classification accuracy of 88.58% and 87.26% for the nearest neighbor classification of remote sensing images in 2020 and 2021 respectively. Meanwhile, the overall classification accuracy for the random forest classification was 92.95% and 92.02% respectively. The Kappa coefficients for the nearest neighbor classification were 0.86 and 0.85, whereas those for the random forest classification were 0.92 and 0.90. These outcomes underscore the utility of the object-oriented classification method in enhancing the accuracy of forest type identification.

Orthogonal binary singular value decomposition method for automated windshield wiper fault detection

For automobile manufacturers, reducing vehicle interior noise is essential for increasing customer satisfaction and vehicle quality. Windshield wipers are one of the major components that generate such noises, and faulty wipers could negatively affect passengers’ psychological and physiological perceptions while driving. Thus, identifying faulty wipers during the manufacturing process would improve the driving experience and vehicle and road safety as well as reduce driver distraction. However, the existing windshield wiper noise-detection process is entirely manual, relies upon human subjectivity, and is time-consuming. Accordingly, this paper develops a novel automated windshield wiper fault-detection system. First, a novel binarization approach is used to effectively binarize the transformed spectrograms of sound signals from windshield wiper operation to segment nAoisy regions. Then, a new matrix-factorisation approach called orthogonal binary singular value decomposition is proposed to decompose binarized mel spectrograms into uncorrelated binary eigenimages to extract meaningful features and identify faulty wipers. Then, the k -nearest neighbour classifier is utilised to classify the extracted features into normal or faulty windshield wipers. Finally, to demonstrate the effectiveness of the proposed system, it was validated on real-life windshield wiper reversal and squeal noise datasets, where it outperformed existing methods and achieved accuracies of 95% and 94%, respectively.

Multi-view local hyperplane nearest neighbor model based on independence criterion for identifying vesicular transport proteins

Vesicular transport proteins participate in various biological processes and play a significant role in the movement of substances within cells. These proteins are associated with numerous human diseases, making their identification particularly important. In this study, we developed a novel strategy for accurately identifying vesicular transport proteins. We developed a novel multi-view classifier called graph-regularized k-local hyperplane distance nearest neighbor model (HSIC-GHKNN), which combines the Hilbert–Schmidt independence criterion (HSIC)-based multi-view learning method with a local hyperplane distance nearest-neighbor classifier. We first extracted protein evolution information using two feature extraction methods, pseudo-position-specific scoring matrix (PsePSSM) and AATP, and addressed dataset imbalance using the Edited Nearest Neighbors (ENN) algorithm. Subsequently, we employed a local hyperplane distance nearest-neighbor classifier for each view identification and added an HSIC term to maintain independence between views. We then assessed the performance of our identification strategy and analyzed the PsePSSM and AATP feature sets to determine the influencing factors of the classification results. The experimental results demonstrate that the accurate and Matthew correlation coefficients of our strategy on the independent test set are 85.8 % and 0.548, respectively. Our approach outperformed existing methods in most evaluation metrics. In addition, the proposed multi-view classification model can easily be applied to similar identification tasks.

Distributed adaptive nearest neighbor classifier: algorithm and theory

Distributed adaptive nearest neighbor classifier: algorithm and theory

An improved intelligent water drops feature selection for finger vein recognition

In recent years, finger vein recognition has gained a lot of attention and been considered as a possible biometric feature. Various feature selection techniques were investigated for intrinsic finger vein recognition on single feature extraction, but their computational cost remains undesirable. However, the retrieved features from the finger vein pattern are massive and include a lot of redundancy. By using fusion methods on feature extraction approaches involving weighted averages, the error rate is minimized to produce an ideal weight. In this research, a novel combinational model of intelligent water droplets is proposed along with hybrid PCA LDA feature extraction for improved finger vein pattern recognition. Initially, finger vein images are pre-processed to remove noise and improve image quality. For feature extraction, Linear Discriminant Analysis (LDA) and Principle Component Analysis (PCA) are employed to identify the most relevant characteristics. The PCA and LDA algorithms combine features to accomplish feature fusion. A global best selection method using intelligent water drops (GBS-IWD) is employed to find the ideal characteristics for vein recognition. The K Nearest Neighbour Classifier was used to recognize finger veins based on the selected optimum features. Based on empirical data, the proposed method decreases the equal error rate by 0.13% in comparison to existing CNN, 3DFM, and JAFVNet techniques. The overall accuracy of the proposed GBSPSO-KNN is 3.89% and 0.85% better than FFF and GWO, whereas, the proposed GBSIWD-KNN is 4.37% and 1.35% better than FFF and GWO respectively.

Tracking Occupant Activities in Autonomous Vehicles Using Capacitive Sensing

Autonomous vehicles (AV) are a promising contemporary engineering innovation. Since they have no human drivers, their vehicle controllers must be fully informed of the activities of their occupants in vehicle seats so that occupant safety and comfort can be guaranteed. This paper introduces a system that employs capacitive sensing and machine learning to inform the controller of occupant activities, like actions and posture changes. The system facilitates capacitive sensing by deploying a capacitance-sensing mat on the vehicle seat. A sensing circuitry connected to the mat measures all its capacitances continuously. Since an occupant’s body induces variations in these capacitances, temporal sequences of capacitance measurements represent various occupant activities in the vehicle seat. Subsequently, the system converts capacitance measurements corresponding to every time step to two grayscale capacitance-sensing images (CSIs). The CSIs, in turn, yield a feature vector corresponding to every time step, thereby building a dataset of temporal sequences of features. The system’s machine-learning unit tracks and recognizes occupant actions and posture changes using an action-recognition block, which is essentially a long short-term memory (LSTM) network trained on a dataset of temporal sequences of either features or capacitance measurements corresponding to various occupant actions. If the occupant remains stationary, a switching block in the unit disables the action-recognition block and enables a posture-recognition block. The latter is a k -nearest-neighbor ( k NN) classifier trained on a dataset of features to recognize stationary occupant postures. This paper validates the system’s performance and investigates its deployment for real-time tracking of occupant activities in AVs.

IARMTS: Design of an Interference-Aware Routing Model with Time Synchronization Capabilities for Dense Wireless Sensor Network Deployments

Performance of dense wireless sensor networks is often degraded due to communication interference and time synchronization issues. Existing machine learning &amp; deep learning models that propose bioinspired &amp; pre-emptive packet-analysis solutions for these tasks either have high complexity, or high deployment costs. Moreover, these models cannot be scaled for heterogeneous node &amp; traffic types, which limits their applicability when applied to real-time scenarios. To overcome these issues, this text proposes design of an interference-aware routing model with time synchronization capabilities for dense wireless sensor network deployments. The network initially collects temporal clock states &amp; packet delivery performance of different nodes on heterogeneous traffic scenarios. These traffic patterns are converted into frequency, entropy, Gabor, and Wavelet components. The converted components are used to train an ensemble set of Naïve Bayes (NB), k Nearest Neighbour (kNN), Multilayer Perceptron (MLP), and Support Vector Machine (SVM) classifiers. These classifiers assist in identification of optimal clock deviations and set of routing paths. These routing paths are further fine-tuned via use of a Bacterial Foraging Optimization (BFO) Model, which assists in identification of interference-aware paths. The BFO Model uses a temporal fitness function that fuses throughput, communication delay, energy levels, and packet delivery performance for different set of contextual communications. Due to which, the model is able to showcase lower end-to-end delay, higher throughput, lower energy consumption, and higher packet delivery performance when compared with existing routing methods under high density nodes &amp; heterogeneous network scenarios. The model showcases 99% PDR, 18.3% lower delay, 19.5% higher energy efficiency and 10.4% lower delay levels when compared with existing methods.

Three-way decision-based noise correction for crowdsourcing

In crowdsourcing learning, a certain degree of label noise still exists in the integrated labels, which are obtained by label integration algorithms. In order to decrease the impact of label noise, many scholars focus on noise correction algorithms. Existing noise correction algorithms usually divide a data set with integrated labels into a clean set and a noise set, directly. However, there are certain noise instances in the clean set and clean instances in the noise set. To reduce the interference from noise instances and improve the utilization of clean instances, inspired by the three-way decision theory, we propose a three-way decision-based noise correction (TDNC) algorithm. Firstly, we calculate the compactness of each instance's multiple noisy label set and estimate the label certainty of each instance's integrated label. Secondly, we divide a data set with integrated labels into three disjoint sets, called positive set, boundary set and negative set, respectively. Thirdly, we further divide and add the instances in the boundary set into the positive set and the negative set according to the nearest neighbor classification. Finally, we train two heterogeneous classifiers on the new positive set to correct the noise instances in the new negative set using a consensus voting strategy. The experimental results on 34 simulated and two real-world data sets consistently show that TDNC significantly outperforms the existing state-of-the-art noise correction algorithms in terms of the noise ratio.

Hybrid machine learning techniques in the management of harmful algal blooms impact

Harmful algal blooms (HABs) are episodes of high concentrations of algae that are potentially toxic for human consumption. Mollusc farming can be affected by HABs because, as filter feeders, they can accumulate high concentrations of marine biotoxins in their tissues. To avoid the risk to human consumption, harvesting is prohibited when toxicity is detected. At present, the closure of production areas is based on expert knowledge and the existence of a predictive model would help when conditions are complex and sampling is not possible. Although the concentration of toxin in meat is the method most commonly used by experts in the control of shellfish production areas, it is rarely used as a target by automatic prediction models. This is largely due to the irregularity of the data due to the established sampling programs. As an alternative, the activity status of production areas has been proposed as a target variable based on whether mollusc meat has a toxicity level below or above the legal limit. This new option is the most similar to the actual functioning of the control of shellfish production areas. For this purpose, we have made a comparison between hybrid machine learning models like Neural-Network-Adding Bootstrap (BAGNET) and Discriminative Nearest Neighbor Classification (SVM-KNN) when estimating the state of production areas. The study has been carried out in several estuaries with different levels of complexity in the episodes of algal blooms to demonstrate the generalization capacity of the models in bloom detection. As a result, we could observe that, with an average recall value of 93.41% and without dropping below 90% in any of the estuaries, BAGNET outperforms the other models both in terms of results and robustness.

P3DC-shot: Prior-driven discrete data calibration for nearest-neighbor few-shot classification

Nearest-Neighbor (NN) classification has been proven as a simple and effective approach for few-shot learning. The query data can be classified efficiently by finding the nearest support class based on features extracted by pre-trained deep models. However, NN-based methods are sensitive to the data distribution and may produce false prediction if the samples in the support set happen to lie around the distribution boundary of different classes. To solve this issue, we present P3DC-shot, an improved nearest-neighbor based few-shot classification method empowered by prior-driven data calibration. Inspired by the distribution calibration technique which utilizes the distribution or statistics of the base classes to calibrate the data for few-shot tasks, we propose a novel discrete data calibration operation which is more suitable for NN-based few-shot classification. Specifically, we treat the prototypes representing each base class as priors and calibrate each support data based on its similarity to different base prototypes. Then, we perform NN classification using these discretely calibrated support data. Results from extensive experiments on various datasets show our efficient non-learning based method can outperform or at least comparable to SOTA methods which need additional learning steps.

Monitoring Software Execution Flow Through Power Consumption and Dynamic Time Warping

This work presents a technique for non-intrusive code execution tracking using side-channel signals of power consumption. Using a nearest-neighbor classifier that integrates the Dynamic Time Warping distance with information from the Control Flow Graph, it is possible to identify executed basic blocks from a trace of power consumption that exhibits temporal distortions due to assembly-level artifacts and varying operational conditions. Experimental results show that the proposed technique achieves over 95% precision when inferring the runtime execution flow of a cruise control application using unmarked traces of power consumption collected from different processors.