K-nearest Neighbor Algorithm Research Articles

Adaptive Invariant Object Representation

In this work, one new approach for RSTB-invariant object representation is presented based on the modified Mellin–Fourier Transform (MFT). For this, in the well-known steps of MFT, the logarithm operation in the log-polar transform is replaced by the operation “rising on a power”. As a result, the central part of the processed area is represented by a significantly larger number of points (transform coefficients), which permits us to give a more accurate description of the main part of the object. The symmetrical properties of the complex conjugated transform coefficients were used, and as a result, the number of coefficients participating in the object representation can be halved without deteriorating the quality of the restored image. The invariant representation is particularly suitable when searching for objects in large databases, which comprise different classes of objects. To verify the performance of the algorithm, object search experiments using the K-Nearest Neighbors (KNN) algorithm were performed, which confirmed this idea. As a result of the analysis, it can be concluded that the complexity of the solutions based on the proposed method depends on the applications, and the inclusion of neural networks is suggested. The neural networks have no conflict with the proposed idea and can only support decision making.

Integration of CNN Models and Machine Learning Methods in Credit Score Classification: 2D Image Transformation and Feature Extraction

Abstract The problem of accurately classifying credit scores is critical for financial institutions to assess individual creditworthiness and effectively manage credit risk. Traditional methods often face limitations when processing large datasets, resulting in lower accuracy and longer processing time. To address this issue, this paper proposes a novel approach to credit score classification by integrating convolutional neural networks (CNN) with machine learning methods. First, a 1D dataset of sequential text data is transformed into 2D greyscale images to use 2D CNN models for feature extraction and classification. Six CNN architectures—DenseNet201, GoogLeNet, MobileNetV2, ResNet18, ShuffleNet, and SqueezeNet—are implemented, and the features in the last layer (1000 features) of each CNN are classified using the softmax method. To further improve the performance, the two best CNN models were selected, and a new fully connected layer (NewFC) was added. A class-based feature set [3 × 31,695] representing three credit score types (good, poor, and standard) was extracted from each model and merged into a feature set [6 × 31,695]. This combined feature set was then reclassified using KNN, LDA, Naive Bayes, and SVM algorithms. The performance of both CNN and machine learning methods was evaluated using accuracy, precision, sensitivity, specificity, and F-score metrics. To optimize classification performance and reduce computational cost, the RelieF algorithm was used to select the best 5 out of 6 features. Compared to using all 6 features, significant improvements in accuracy and efficiency were observed, demonstrating the effectiveness of the proposed method in credit score classification.

Classification of Pear Varieties Using the K-Nearest Neighbor Algorithm and Extraction of Shape, Color, Texture, and Size Features

This study develops a pear variety classification system based on digital images using the K-Nearest Neighbor (KNN) algorithm. The data used included 195 images from three pear varieties, namely Century, Forel Afrika, and Singo, which were analyzed by utilizing various features such as color (RGB), texture (Local Binary Pattern), shape (area, circumference, length-width ratio), and size (bounding box dimensions). The preprocessing process removes the image's background to increase focus on the main object, thus allowing for more optimal feature extraction. The dataset is divided into 80% for training and 20% for model testing. The evaluation results show that the KNN model can achieve an accuracy of 85%, with an average precision value of 0.85, recall of 0.89, and F1-score of 0.85. These results prove that the KNN algorithm is effective in accurately classifying pear varieties, which can significantly contribute to applying digital image-based technology for automatic classification needs in the agricultural sector.

Influence of Age on Second Language Acquisition

Age is one of the important factors affecting individual differences in second language acquisition. The development of cognitive ability also has certain influence on second language acquisition, depending on whether this influence is positive or negative. This paper discusses the educational significance of the age factor in English teaching language learning to second language learners based on BD(big data) environment, optimizes the teaching mode of English wisdom class and designs it in detail according to different classes, combining with the related theories of wisdom class and its teaching mode research. Parallelize the improved KNN(k nearest neighbour) algorithm, so that the algorithm can well adapt to the data processing model of cloud computing platform. After parallelization, the improved KNN algorithm runs on the Spark cloud platform, which can quickly improve the efficiency of the algorithm by about 10.3%. The improved KNN algorithm greatly improves the efficiency of the algorithm on the premise of ensuring the classification accuracy of the algorithm.

Enhancing internet of things attack detection using principal component analysis and kernel principal component analysis with cosine distance and sigmoid kernel

The widespread adoption of internet of things (IoT) devices has brought about unprecedented levels of connectivity and convenience. However, it has also introduced significant challenges, particularly in the areas of security and privacy. This study addresses the critical issue of intrusion detection within IoT environments, with a specific focus on analyzing the Iot-23 dataset. Our methodology involves employing principal component analysis (PCA) and kernel PCA for dimensionality reduction. Subsequently, we utilize the k-nearest neighbors (KNN) algorithm for classification purposes. To optimize the performance of the KNN algorithm, we experiment with various feature scaling techniques, such as StandardScaler, MinMaxScaler, and RobustScaler, utilizing different distance metrics. In our analysis, we discovered that employing the cosine distance metric in combination with KNN resulted in superior intrusion detection performance when utilizing PCA. Additionally, when utilizing kernel PCA, we evaluated multiple kernel functions and determined that the radial basis function and sigmoid kernel yielded the most favorable results.

E-commerce Service Chatbot Application Design using KNN and Random Forest Methods

Our lives are profoundly shaped by technology, with the expansion of e-commerce being a notable outcome that demands attention. Given the prevalence of smartphones equipped with rapid messaging and networking applications, individuals often utilize these platforms to communicate with sellers, offering a convenient means for sellers to efficiently engage with a diverse customer base. Recognizing this trend, there arises a necessity for the digital transformation of services to enhance operational efficiency. In response to this need, the researcher has developed a chatbot application aimed at improving customer service, employing machine learning techniques with the KNN and Random Forest algorithms. To assess the application's viability, the chatbot's results undergo an accuracy test, revealing a satisfactory accuracy value of 71.4%, thereby affirming its feasibility.

Prediction of Beef Production Using Linear Regression, Random Forest and k-Nearest Neighbors Algorithms

The rapid increase in the global population and evolving dietary habits have significantly heightened the demand for high-quality protein sources. Beef, as a vital protein source, plays a crucial role in meeting this growing demand. This study aims to develop and evaluate a machine-learning model to predict beef production using meteorological, agricultural, and economic data. To achieve this, three different machine learning algorithms—Linear Regression, Random Forest, and k-Nearest Neighbors—were employed. The results indicate that the Random Forest algorithm outperformed the other methods in terms of R² and error metrics, demonstrating superior predictive accuracy. The study highlights the potential of machine learning techniques in predicting beef production, offering valuable insights for stakeholders involved in strategic decision-making to meet nutritional needs. As the global demand for protein continues to rise, the importance of such predictive models becomes increasingly significant, emphasizing the distinct advantages that machine learning approaches provide in this context.

Identification of novel serum lipid metabolism potential markers and metabolic pathways for oral cancer: a population-based study

ObjectiveThis study aims to identify potential lipid biomarkers and metabolic pathways associated with oral cancer (OC). Then to establish and evaluate disease classification models capable of distinguishing OC patients from healthy controls.MethodsA total of 41 OC patients and 41 controls were recruited from a hospital in Southeast China to examine the serum lipidomics by Ultra-high Performance Liquid Chromatography Q Exactive Mass Spectrometry (UHPLC-QE-MS).ResultsThe total serum lipid profile showed that triglycerides accounted for the highest proportion of total metabolites, reaching 35.90% of the total. A total of 74 different metabolites were screened (12 up-regulated and 62 down-regulated), mainly enriched in the glycerophospholipid metabolism pathway. The three most significant changes in lipid metabolites were phosphatidylcholine (PC(18:3e/17:2)), acylcarnitine (ACar(14:2)), and glucuronosyldiacylglycerol (GlcADG(14:1/14:1)). The disease classification model, constructed using a KNN algorithm with 13 metabolites selected through LASSO screening, achieved the best performance, with an AUC of 0.978 (0.955-1.000).ConclusionLipid metabolic biomarkers identified in this study exhibit potential as candidate biomarkers for OC diagnosis. Further validation through prospective studies is required to confirm their clinical utility in early detection.

A hybrid deep learning air pollution prediction approach based on neighborhood selection and spatio-temporal attention

Air pollution is a critical global environmental issue, further exacerbated by rapid industrialization and urbanization. Accurate prediction of air pollutant concentrations is essential for effective pollution prevention and control measures. The complex nature of pollutant data is influenced by fluctuating meteorological conditions, diverse pollution sources, and propagation processes, underscores the crucial importance of the spatial and temporal feature extraction for accurately predicting air pollutant concentrations. To address the challenges of data redundancy and diminished long-term prediction accuracy observed in previous studies, this paper presents an innovative approach to predict air pollutant concentrations leveraging advanced data analysis and deep learning methods. The proposed approach, termed KSC-ConvLSTM, integrates the k-nearest neighbors (KNN) algorithm, spatio-temporal attention (STA) mechanism, the residual block, and convolutional long short-term memory (ConvLSTM) neural network. The KNN algorithm adaptively selects highly correlated neighboring domains, while the residual block, enhanced with the STA mechanism, extracts spatial features from the input data. ConvLSTM further processes the output from STA-ConvNet to capture high-dimensional temporal and spatial features. The effectiveness of the KSC-ConvLSTM approach was validated through predictions of PM2.5 concentrations in Beijing and its surrounding urban agglomeration. The experimental results indicate that the KSC-ConvLSTM approach outperforms benchmark approaches in single-step, multi-step, and trend prediction. It demonstrates superior fitting accuracy and predictive performance. Quantitatively, the proposed KSC-ConvLSTM approach reduces the root mean square error (RMSE) by 4.216–8.458 for prediction averages of 1–12 h of PM2.5 in Beijing, compared with the benchmark approach. The findings show that the KSC-ConvLSTM approach shows considerable potential for predicting, preventing, and controlling air pollution.

Aircraft Sensor Fault Diagnosis Based on GraphSage and Attention Mechanism

Aircraft sensors are crucial for ensuring the safe and efficient operation of aircraft. However, these sensors are vulnerable to external factors that can lead to malfunctions, making fault diagnosis essential. Traditional deep learning-based fault diagnosis methods often face challenges, such as limited data representation and insufficient feature extraction. To address these problems, this paper proposes an enhanced GraphSage-based fault diagnosis method that incorporates attention mechanisms. First, signal data representing the coupling characteristics of various sensors are constructed through data stacking. These signals are then transformed into graph data with a specific topology reflecting the overall sensor status of the aircraft using K-nearest neighbor and Radius classification algorithms. This approach helps fully leverage the correlations between data points. Next, node and neighbor information is aggregated through graph sampling and attention-based aggregation methods, strengthening the extraction of fault features. Finally, fault diagnosis is performed using multi-layer aggregation and transformation within fully connected layers. Experiments demonstrate that the proposed method outperforms baseline approaches, achieving better detection performance and faster computational speed. The method has been validated on both simulated and real-flight data.

K-nearest neighbor algorithm for imputing missing longitudinal prenatal alcohol data

AimsThe objective of this study is to illustrate the application of a machine learning algorithm, K Nearest Neighbor (k-NN) to impute missing alcohol data in a prospective study among pregnant women.MethodsWe used data from the Safe Passage study (n = 11,083). Daily alcohol consumption for the last reported drinking day and 30 days prior was recorded using the Timeline Follow back method, which generated a variable amount of missing data per participants. Of the 3.2 million person-days of observation, data were missing for 0.36 million (11.4%). Using the k-NN imputed values were weighted for the distances and matched for the day of the week. Since participants with no missing days were not comparable to those with missing data, segments of non-missing data from all participants were included as a reference. Validation was done after randomly deleting data for 5–15 consecutive days from the first trimester.ResultsWe found that data from 5 nearest neighbors (i.e., K = 5) and segments of 55 days provided imputed values with least imputation error. After deleting data segments from the first trimester data set with no missing days, there was no difference between actual and predicted values for 64% of deleted segments. For 31% of the segments, imputed data were within +/−1 drink/day of the actual. Imputation accuracy varied by study site because of the differences in the magnitude of drinking and proportion of missing data.Conclusionk-NN can be used to impute missing data from longitudinal studies of alcohol during pregnancy with high accuracy.

Algorithmic Trading Bot Using Artificial Intelligence Supertrend Strategy

Abstract: This article presents a trading strategy that combines the Super Trend indicator with the K-Nearest Neighbors (KNN) algorithm, utilizing artificial intelligence (AI) to automate market decision-making and enhance trading accuracy. The strategy integrates the SuperTrend indicator, which dynamically tracks market volatility, with the KNN algorithm, allowing the system to classify market trends as bullish, bearish, or neutral based on historical data. This enables the strategy to make intelligent, data-driven decisions in real-time without human intervention. The AI-driven approach automates the entire trading process, from data analysis to trade execution, improving efficiency and removing emotional biases from trading decisions. The KNN algorithm plays a key role in this automation by analyzing past market conditions and identifying patterns that inform future price movements. This allows the system to adapt to changing market trends and react quickly to new data, ensuring timely and accurate decisions. The results of the strategy indicate strong performance, with a Net Profit of 959.38 USD and a Gross Profit of 3,005.71 USD, demonstrating the strategy's ability to generate consistent returns. The Profit Factor of 1.469 further highlights the system's ability to produce profits while managing risk effectively. Additionally, the Sharpe Ratio of 0.558 shows that the strategy provides positive risk-adjusted returns, making it a reliable tool for automated trading. In conclusion, this AI-powered Super Trend-KNN strategy showcases the potential of combining artificial intelligence with technical indicators for automated trading. By eliminating the need for manual intervention and leveraging AI to adapt to market conditions, the strategy provides an efficient and scalable solution for intelligent decision-making in trading.

Non-Invasive and Accurate Blood Glucose Detection Based on an Equivalent Bioimpedance Spectrum

Accurate blood glucose monitoring is a key issue for the diagnosis and treatment of diabetes. For this purpose, a non-invasive blood glucose detection method is proposed, which makes use of the equivalent bioelectric impedance spectrum. An impedance detection platform is designed using an automatic balance bridge technique, which can acquire an impedance spectrum within the range of dispersion. Then, the K-nearest neighbor algorithm is used to extract the characteristics of the impedance spectrum. Furthermore, higher-order multiple regression methods are used to establish a blood glucose–electrical impedance spectrum model. Experimental results show that the proposed blood glucose–electrical impedance spectrum model can estimate the change in blood glucose and reliably identify the high blood glucose samples. The correlation between the proposed method and the biochemical blood glucose values can reach 0.89 and 0.87 in personal and multi-person blood glucose tests, respectively. Thus, the proposed method provides a feasible solution for non-invasive blood glucose detection and can help us identify diabetes mellitus.

An Artificial Intelligence Home Monitoring System That Uses CNN and LSTM and is Based on the Android Studio Development Platform

This paper developed an artificial intelligence home environment monitoring system by using the Android Studio development platform. A database was constructed within a server to store sensor data. The proposed system comprises multiple sensors, a message queueing telemetry transport (MQTT) communication protocol, cloud data storage and computation, and end device control. A mobile application was developed using MongoDB software, which is a file-oriented NoSQL database management system developed using C++. This system represents a new database for processing big sensor data. The k-nearest neighbor (KNN) algorithm was used to impute missing data. Node-RED development software was used within the server as a data-receiving, storage, and computing environment that is convenient to manage and maintain. Data on indoor temperature, humidity, and carbon dioxide concentrations are transmitted to a mobile phone application through the MQTT communication protocol for real-time display and monitoring. The system can control a fan or warning light through the mobile application to maintain ambient temperature inside the house and to warn users of emergencies. A long short-term memory (LSTM) model and a convolutional neural network (CNN) model were used to predict indoor temperature, humidity, and carbon dioxide concentrations. Average relative errors in the predicted values of humidity and carbon dioxide concentration were approximately 0.0415% and 0.134%, respectively, for data storage using the KNN algorithm. For indoor temperature prediction, the LSTM model had a mean absolute percentage error of 0.180% and a root-mean-squared error of 0.042 °C. The CNN–LSTM model had a mean absolute percentage error of 1.370% and a root-mean-squared error of 0.117 °C.

Online forecasting using neighbor-based incremental learning for electricity markets

Abstract Electricity market forecasting is very useful for the different actors involved in the energy sector to plan both the supply chain and market operation. Nowadays, energy demand data are data coming from smart meters and have to be processed in real-time for more efficient demand management. In addition, electricity prices data can present changes over time such as new patterns and new trends. Therefore, real-time forecasting algorithms for both demand and prices have to adapt and adjust to online data in order to provide timely and accurate responses. This work presents a new algorithm for electricity demand and prices forecasting in real-time. The proposed algorithm generates a prediction model based on the k-nearest neighbors algorithm, which is incrementally updated in an online scenario considering both changes to existing patterns and adding new detected patterns to the model. Both time-frequency and error threshold based model updates have been evaluated. Results using energy demand from 2007 to 2016 and prices data for different time periods from the Spanish electricity market are reported and compared with other benchmark algorithms.

Random k conditional nearest neighbor for high-dimensional data.

The k nearest neighbor (kNN) approach is a simple and effective algorithm for classification and a number of variants have been proposed based on the kNN algorithm. One of the limitations of kNN is that the method may be less effective when data contains many noisy features due to their non-informative influence in calculating distance. Additionally, information derived from nearest neighbors may be less meaningful in high-dimensional data. To address the limitation of nearest-neighbor based approaches in high-dimensional data, we propose to extend the k conditional nearest neighbor (kCNN) method which is an effective variant of kNN. The proposed approach aggregates multiple kCNN classifiers, each constructed from a randomly sampled feature subset. We also develop a score metric to weigh individual classifiers based on the level of separation of the feature subsets. We investigate the properties of the proposed method using simulation. Moreover, the experiments on gene expression datasets show that the proposed method is promising in terms of predictive classification performance.

WALLABY Pilot Survey: kNN identification of perturbed galaxies through Hi morphometrics

Abstract Galaxy morphology in stellar light can be described by a series of “non-parametric” or “morphometric” parameters, such as concentrationasymmetry-smoothness, Gini, M20, and Sersic fit. These parameters can be applied to column density maps of atomic hydrogen (Hi). The Hi distribution is susceptible to perturbations by environmental effects, e.g. inter-galactic medium pressure and tidal interactions. Therefore, Hi morphology can potentially identify galaxies undergoing ram-pressure stripping or tidal interactions. We explore three fields in theWALLABY Pilot Hi survey and identify perturbed galaxies based on a k-nearest Neighbor (kNN) algorithm using an Hi morphometric feature space. For training, we used labeled galaxies in the combined NGC 4808 and NGC 4636 fields with six Hi morphometrics to train and test a kNN classifier. The kNN classification is proficient in classifying perturbed galaxies with all metrics –accuracy, precision and recall– at 70-80%. By using the kNN method to identify perturbed galaxies in the deployment field, the NGC 5044 mosaic, we find that in most regards, the scaling relations of perturbed and unperturbed galaxies have similar distribution in the scaling relations of stellar mass vs star formation rate and the Baryonic Tully-Fisher relation, but the Hi and stellar mass relation flatter than of the unperturbed galaxies. Our results for NGC 5044 provide a prediction for future studies on the fraction of galaxies undergoing interaction in this catalogue and to build a training sample to classify such galaxies in the fullWALLABY survey.

Rapid identification of marine microplastics by laser-induced fluorescence technique based on PCA combined with SVM and KNN algorithm.

The laser-induced fluorescence technique has the advantage of fast and non-destructive detection and can be used to classify types of marine microplastics. However, spectral overlap poses a challenge for qualitative and quantitative analysis by conventional fluorescence spectroscopy. In this paper, a 405 nm excitation laser source was used to irradiate 4 types of microplastic samples with different concentrations, and a total of 1600 sets of fluorescence spectral data were obtained. The 726 data points contained in each sample spectrum were first analyzed by PCA, and the 4 microplastics were differentiated by their position in the PCA score plot. The classification and identification are then performed by SVM, KNN, PCA-SVM and PCA-KNN algorithms respectively. The classification accuracy of microplastics in seawater using SVM and KNN algorithms is higher than 86%. The classification accuracy can be increased to 100% by PCA combined with SVM and KNN algorithm. Concentration inversion was conducted by SVM and KNN algorithms after classification. The correlation coefficients between the predicted values and the actual values were higher than 0.8, and the RMSE was less than 0.47%, which indicated that both algorithms had good prediction results. These machine learning methods provide accurate and reliable identification results in the rapid identification of microplastic types and their concentrations without complex spectral data preprocessing and fluorescence background removal algorithms.

Monkeypox diagnosis based on probabilistic K-nearest neighbors (PKNN) algorithm.

Although it is not a new illness and has been around since the previous century, monkeypox later resurgence is fraught with difficulties. This study presents a novel approach of diagnosing monkeypox using artificial intelligence, which is called Effective Monkeypox Diagnosis Strategy (EMDS). The proposed EMDS is established through two sequential stages, namely; (i) Pre-Processing Phase (PP) and (ii) Monkeypox Diagnosing phase (MDP). During PP the input image dataset is prepared through three processes, which are; feature extraction, feature selection, and anomaly rejection, while the actual diagnosis performed in the MDP. Features are extracted from input images using GoogleNet as an effective pre-trained deep learning model, while Leopard Seal Optimization (LSO) is employed to select the most instructive features. The major contribution of this paper is focused in two issues, which are; (i) introducing a new methodology for rejecting anomalies from the input image dataset based on interquartile range (IQR), and (ii) proposing a new instance of K-Nearest Neighbor classifier for monkeypox diagnosis, which is called; Probabilistic K-Nearest Neighbors (PKNN) Algorithm. The proposed PKNN combines evidence from distance based traditional KNN as well as the Naïve probabilistic theorem used by Naïve Bayes (NB) algorithm in an integrated way. Numerous experiments have been conducted considering the proposed EMDS as well as recent competitive strategies on two public monkeypox datasets, which are; the Monkeypox Skin Image and Lesion Datasets (MSID and MSLD, respectively). Initially, the performance of the basic contributions of the proposed EMDS, which are the outlier rejection methodology (ORM) and PKNN, are evaluated individually. Then, EMSD as a whole is evaluated. Moreover, an ablation study has also been conducted to evaluate the effect of ORM and PKNN on the performance of EMSD. Based on the experimental results, it is shown that EMDS outperforms recent monkeypox identification strategies as it achieves 99% diagnosis accuracy. Moreover, it indicates the maximum precision and recall with the minimum diagnosis time.

Reduction in Microgrid Topology Selection Time via Hybrid Branch and Bound and k-Nearest Neighbors Techniques

The global adoption of hybrid renewable energy systems (HRESs) is accelerating as a strategic response to escalating energy demands and the imperative to mitigate greenhouse gas emissions. Despite the development of various technological tools, such as pre-feasibility analysis, sizing, and simulation tools, challenges persist due to their limited flexibility in modifying system architectures and their typically long computation times, which hinder their practical efficiency. This study introduces a novel hybrid method that integrates the Branch and Bound (BB) heuristic search algorithm with the k-Nearest Neighbors (kNN) algorithm to drastically reduce the simulation time of microgrid models in Simulink. Validation considering four distinct case studies reveals that our method can decrease the simulation time by up to 94.68% while maintaining an acceptable accuracy. Specifically, simulation times in certain cases were reduced from approximately 21,780 and 118,580 s to 1442.7969 and 6306.0625 s, respectively. This significant reduction facilitates the rapid evaluation and selection of optimal HRES configurations, enhancing the efficiency of both editable and non-editable systems. Through streamlining the simulation process, this approach not only accelerates the design and analysis phases but also supports the broader adoption and deployment of HRESs, which is critical for achieving a sustainable future. This advancement offers a robust and efficient methodology for optimizing simulation times, thereby addressing a key bottleneck in the development and implementation of hybrid renewable energy solutions.