K-nearest Neighbor Technique Research Articles

Safe Data-Driven Lane Change Decision Using Machine Learning in Vehicular Networks

This research proposes a unique platform for lane change assistance for generating data-driven lane change (LC) decisions in vehicular networks. The goal is to reduce the frequency of emergency braking, the rate of vehicle collisions, and the amount of time spent in risky lanes. In order to analyze and mine the massive amounts of data, our platform uses effective Machine Learning (ML) techniques to forecast collisions and advise the driver to safely change lanes. From the unprocessed large data generated by the car sensors, kinematic information is retrieved, cleaned, and evaluated. Machine learning algorithms analyze this kinematic data and provide an action: either stay in lane or change lanes to the left or right. The model is trained using the ML techniques K-Nearest Neighbor, Artificial Neural Network, and Deep Reinforcement Learning based on a set of training data and focus on predicting driver actions. The proposed solution is validated via extensive simulations using a microscopic car-following mobility model, coupled with an accurate mathematical modelling. Performance analysis show that KNN yields up to best performance parameters. Finally, we draw conclusions for road safety stakeholders to adopt the safer technique to lane change maneuver.

A TWO-TIER MACHINE LEARNING FRAMEWORK FOR RISK ASSESSMENT IN DRIVERS WITH CARDIOVASCULAR DISORDERS

This work proposes a scheme based on a two-tier machine learning (ML) framework for the initial screening of commercial drivers with cardiovascular disorders prior to the actual driving assessment. First, the proposed framework aims to provide primary health care to cardiac drivers in resource-constrained scenarios such as bus terminals with the help of paramedical staff. The prediction of cardiovascular disease (CVD) in drivers is done using a variety of ML approaches, including Support Vector Machines (SVMs), Random Forests (RFs), Logistic Regression (LR), K-Nearest Neighbor (KNN), Decision Trees (DT), Naive Bayes (NB), and XG-Boost (XGB). The K-fold cross-validation technique also tests the model’s ability to predict CVD. Second, a no-drive alert will be provided whenever the model predicts heart disease, and a comma-separated value (CSV) file stores the predicted abnormal parameters. An email-based data communication has been set up to transfer the CSV file. A MySQL database has been created to store the abnormal data received in hospitals which will help cardiologists with the proper diagnosis. This internet of medical things (IoMT) process will enable divers to come to the hospital for medication only when advised by a cardiologist, thereby reducing the burden of routine hospital visits. The Cleveland database of the UCI ML repository, a multivariate CVD dataset that contains 14 features from 303 people, is utilized to test the performance of the proposed model. Also, the proposed model performance is evaluated using two more publicly available heart disease datasets, i.e., the MIT-BIH arrhythmia dataset and the CVD dataset. The XGB, KNN, and RF ML techniques outperform state-of-the-art methods with performance accuracies of 88.53%, 91.8%, and 93.44%, respectively, for the Cleveland database; performance accuracies of 99.20%, 98.82%, and 99.08% for the MIT-BIH arrhythmia dataset; and performance accuracies of 73.29%, 69.48%, and 71.74% for the CVD dataset. Furthermore, the results showed comparable performance to the rest of the ML techniques. Early detection of CVD and consultation with specialist doctors are essential before it reaches a seriousness that can save drivers from vehicular accidents while seeking health care.

Machine learning based performance estimation of terrestrial Ro-FSO Link

In this work, a radio over free-space optical communication (Ro-FSO) link has been examined considering quadrature amplitude modulation (64-QAM) based orthogonal frequency division multiplexing (OFDM) technique for a turbulence channel. The performance of the system has been investigated considering log normal and gamma-gamma atmospheric scintillation models under clear air, rain and fog weather conditions. Artificial neural network (ANN), k-nearest neighbour (KNN), and decision tree (DT) machine learning (ML) techniques have been applied for estimation of quality of received signal in terms of bit error rate BER. ANN model exhibits the highest value of R-squared (R2) of 0.9967 and lowest value of root mean square error (RMSE) of 0.0134 as compared to other ML techniques resulting in the best fit model.

Evaluation of IoT-Based Smart Home Assistance for Elderly People Using Robot

In the development of Internet-of-things (IoT)-based technology, there is a pre-programmed robot called Cyborg which is used for assisting elderly people. It moves around the home and observes the surrounding conditions. The Cyborg is developed and used in the smart home system. The features of a smart home system with IoT technology include temperature control, lighting control, surveillance, security, smart electricity, and water sensors. Nowadays, elderly people may not be able to maintain their homes appropriately and may feel uncomfortable performing day-to-day activities. Therefore, Cyborg can be used to carry out the activities of elderly people. Such activities include switching off unnecessary lights, watering plants, gas control, monitoring intruders or unknown persons, alerting elderly people in emergency situations, etc. These activities are controlled by web-based platforms as well as smartphone applications. The issues with the existing algorithms include that they are inefficient, require a long time for implementation, and have high storage space requirements. This paper proposes the k-nearest neighbors (KNN) with an artificial bee colony (ABC) algorithm (KNN-ABC). In this proposed work, KNN-ABC is used with wireless sensor devices for perceiving the surroundings of the smart home. This work implements the automatic control of electronic appliances, alert signal processors, intruder detection, and performs day-to-day activities automatically in an efficient way. GNB for intruder detection in the smart home environment system using the Cyborg human intervention robot achieved an accuracy rate of 88.12%, the Artificial Bee Colony algorithm (ABC) achieved 90.12% accuracy on the task of power saving in smart home electronic appliances, the KNN technique achieved 91.45% accuracy on the task of providing a safer pace to the elderly in the smart home environment system, and our proposed KNN-ABC system achieved 93.72%.

A new multivariate control chart based on the isolation forest algorithm

Statistical process control (SPC) is an extensively utilized technique in quality management for service and manufacturing industries. Traditional SPC control charts rely on normality assumptions and analyze data in one dimension. Machine learning methods have been increasingly used to address limitations of traditional SPC techniques, such as one-class classification (OCC) procedures, using support vector machine (SVM), K-nearest neighbor (KNN), and other techniques. Nonetheless, there are drawbacks associated with these methods. For instance, SVM-based approaches may require significant computational resources, whereas KNN-based approaches may encounter difficulties in identifying minor variations. Here, these limitations are addressed using a new SPC framework that incorporates the isolation forest (iForest) algorithm, an unsupervised anomaly detection technique developed for machine learning applications. Our presented method examines how the number of trees impacts the charting scheme and offers several advantages: (1) it has a low constant time complexity that results in fast computing speeds; (2) it outperforms competitors in detecting small shifts in various distributions; and (3) while dealing with challenging distributions with heavy tails, it proves to be effective. We provide two examples that demonstrate the performance of our proposed control chart using Monte Carlo simulations.

English

Researchers and data analysts commonly experience with missing data in their field of studies. It is necessary to handle missing data properly to obtain better and more reliable outcomes of any research. The objective of this research is to evaluate different imputation techniques for handling missing observations in weather data. For this purpose weather data of daily rainfall maximum temperature (Tmax) and minimum temperature (Tmin) of 23 stations of Pakistan were taken from Pakistan Metrological department from 1981 to 2020. There are total 14610 observation of each variable and each variable have different number of missing observations at different stations. For the estimation of missing observation at each station mean imputation k nearest neighbors (KNN) imputation predictive mean matching (PMM) imputation and sample imputation techniques are used. Among different imputation techniques the technique which provides minimum root mean square error (RMSE) is selected and used to estimate the missing observations at each station. The result shows that the selection of imputation method is different for each station and mainly depends upon the size of missing observation in the data. For rainfall data the KNN technique is the most appropriate while for Tmax and Tmin the mean imputation technique is recommended for the estimation of missing observation in the data as compared to other methods.

Detection of power transmission lines faults based on voltages and currents values using K-nearest neighbors

The critical factors to consider when implementing a maintenance plan for energy transmission lines are, accuracy, speed, and time, because of the increased global demand for electricity power caused by rapid development, and overuse of electric power transmission lines (both underground cables and overhead transmission lines), which in turn reduces the efficiency of the lines. Consequently, the efficiency of the lines may be reduced as a result of overuse or other activities like excavation that may have tampered with the cables. Thus, it becomes important to investigate the faults to which the lines are exposed. To this end, this article focuses on the detection of fault in transmission lines through the use of k-nearest neighbor algorithm. Using this algorithm, the characteristics were obtained (voltage, current), and these characteristics enable the identification of faults in the transmission lines, and in the specific location (the entire system, phase B, and phase A). The benefits that can be derived from the use of this algorithm include time, accuracy, speed, which are the requirements for the maintenance of transmission lines. Euclidean distance used in the application of the k-nearest neighbor technique for weights, and K = 3 for number of neighbors. The dataset was split into two parts, 70% training set and 30% testing set.

Climate change forecasting using data mining algorithms

Abstract Water management is very important for human life sustainability. Rainfall forecasting is one of the most important factors for the water management of an area. A forecast is simply a calculation of what happens in the future based on past information under the assumption that the pattern followed in the past would continue in the future also. This work aims at obtaining forecasting models for the time series data set using conventional models and computational models. Varanasi City's annual climate data for a total of 113 years is used for the analysis. Initially, the individual model is considered and used for forecasting. Later, hybrid models will be considered and a comparison between individual models and hybrid models would be obtained. The individual statistical models to be considered are moving average, exponential smoothing with one parameter, and autoregressive integrated moving average (ARIMA). The forecast is also done individually using the k-nearest neighbor (kNN) and interpolation technique cubic spline. Finally, the best-chosen statistical models and the interpolation model are coupled with kNN to develop hybrid models and with these hybrid models, the forecast is done for the data. All the models will be compared and the best among them will be chosen.

Evaluating climatic change effects on EC and runoff in the near future (2020–2059) and far future (2060–2099) in arid and semi-arid watersheds

This study investigates the effect of climate change on the runoff and electrical conductivity (EC) of the Marun watershed. It used 35 general circulation models (GCMs) and the identification of unit hydrographs and component flows from rainfall, evaporation and streamflow data (IHACRES) rainfall-runoff model for the hydrological simulation. Moreover, a non-parametric regression model based on the multivariate adaptive regression splines (MARS) was utilized to estimate the EC under representative concentration pathway RCP4.5 and RCP8.5 scenarios in the near future F1 (2020–2059) and far future F2 (2060–2099) periods. Also, it used the technique for order of preference by similarity to ideal solution (TOPSIS) method to determine the best GCMs for each region and the k-nearest neighbors (KNN) technique to combine the temperature (Tmean) and precipitation (PCP) outputs and reduce the GCM uncertainty in each cell. According to the results, the highest increase of EC relative to the historical period (1966–2005) that will occur in the F1 period under the RCP4.5 and RCP8.5 scenarios is 17.43% and 15.6%, and for the F2 period is 18.46% and 11.2%, respectively, during autumn. The changes of annual Tmean, PCP, runoff, and EC in F1 period are 8.6%, 2.1%, − 10.7%, and − 11%, respectively, under the RCP4.5 scenario and 10.5%, 5.9%, − 3.5%, and − 12.2%, respectively, under the RCP8.5 scenario. The same values for the F2 period are 12.9%, − 0.1%, − 14.9%, and − 10%, respectively, under the RCP4.5 scenario and 22.6%, 5.2%, 1.2%, and − 12.8%, respectively, under the RCP8.5 scenario relative to the historical period.

Machine learning based Software Fault Prediction models

The study aims to identify soft-computing-based software fault prediction models that assist in resolving issues related to the quality, reliability, and cost of the software projects. It proposes models for implementation of software fault prediction using decision-tree regression and the K-nearest neighbor technique of machine learning. The proposed models have been designed and implemented in Python using designed metric suites as input, and the predicted-faults as output, for the real-time, wider dataset from the Promise repository. By comparing the prediction and validation results of the proposed models for the same dataset, it has been concluded that the decision-tree regression-based fault prediction model has the best performance with values of MMRE, RMSE, and accuracy of 0.0000204, 3.54, and 99.37, respectively.

Evaluation of Machine Learning Techniques Towards Early Detection of Cardiovascular Diseases

The effectiveness of three Machine Learning (ML) algorithms: Support Vector Machine (SVM), Random Forest (RF) and K-Nearest Neighbour (KNN) techniques for the early diagnosis of heart diseases were evaluated. Heart disease’ dataset collected from kaggle.com data repository, which comprised of 303 data points with 13 features and a target variable were used and data preprocessing by data shuffling and dimension reduction were performed. The new dimension of the dataset was chosen such that 85.03% of the original information is retained. The preprocessed dataset was partitioned into 70% of the training set and 30% of the testing set. The ML algorithms were trained and tested for the diagnosis of cardiovascular diseases (CVD). The training performances of these models were evaluated with a k-fold cross-validation algorithm using 10 folds. The k-fold accuracy shows KNN with an accuracy of 0.837662, RF with an accuracy of 0.834091, and SVM with an accuracy of 0.814935. The test results also show KNN with an accuracy of 0.8, SVM with an accuracy of 0.7889, and RF with an accuracy of 0.7667. KNN emerged the best model both in training and test’s performances and is recommended for the early diagnosis of CVD.

PL-kNN: A Python-based implementation of a parameterless k-Nearest Neighbors classifier

This paper presents an open-source implementation of PL-kNN, a parameterless version of the k-Nearest Neighbors algorithm. The proposed model, developed in Python 3.6, was designed to avoid the choice of the k parameter required by the standard k-Nearest Neighbors technique. Essentially, the model computes the number of nearest neighbors of a target sample using the data distribution of the training set. The source code provides functions resembling the Scikit-learn methods for fitting the model and predicting the classes of the new samples. The source code is available in the GitHub repository with instructions for installation and examples for usage.

Adaptation strategies for future coastal flooding: Performance evaluation of green and grey infrastructure in South Korea

Climate change is contributing to an increasing frequency and intensity of floods in Korea. This study predicts areas with a high probability of flooding in coastal areas of South Korea owing to future climate change, which is likely to cause extreme rainfall and sea-level rise, using a spatiotemporal downscaled future climate change scenario with random forest, artificial neural network, and k-nearest neighbor techniques. In addition, the change in coastal flooding risk probability according to the application of different adaptation strategies (green spaces and seawalls) was identified. The results showed a clear difference in the risk probability distribution in the absence and presence of either adaptation strategy. Their effectiveness in moderating future flooding risks is subject to change owing to strategy type, geographic region, and urbanization intensity and the results show that green spaces are slightly more effective than seawalls when forecasting for 2050. This demonstrates the importance of a nature-based strategy. Moreover, this study highlights the need to prepare adaptation measures according to regional characteristics to mitigate the impact of climate change. Korea is surrounded by seas on three sides that have independent geophysical and climate characteristics. The south coast has a higher risk of coastal flooding than the east and west coasts. In addition, a higher urbanization rate is associated with a higher risk probability. This implies that climate change response strategies for coastal cities are necessary as the population and socioeconomic activities of coastal urban areas are likely to increase in the future.

Semi-Supervised Multi-Label Dimensionality Reduction Learning by Instance and Label Correlations

The label learning mechanism is challenging to integrate into the training model of the multi-label feature space dimensionality reduction problem, making the current multi-label dimensionality reduction methods primarily supervision modes. Many methods only focus attention on label correlations and ignore the instance interrelations between the original feature space and low dimensional space. Additionally, very few techniques consider how to constrain the projection matrix to identify specific and common features in the feature space. In this paper, we propose a new approach of semi-supervised multi-label dimensionality reduction learning by instance and label correlations (SMDR-IC, in short). Firstly, we reformulate MDDM which incorporates label correlations as a least-squares problem so that the label propagation mechanism can be effectively embedded into the model. Secondly, we investigate instance correlations using the k-nearest neighbor technique, and then present the l1-norm and l2,1-norm regularization terms to identify the specific and common features of the feature space. Experiments on the massive public multi-label data sets show that SMDR-IC has better performance than other related multi-label dimensionality reduction methods.

Big data classification based on improved parallel k-nearest neighbor

In response to the rapid growth of many sorts of information, highway data has continued to evolve in the direction of big data in terms of scale, type, and structure, exhibiting characteristics of multi-source heterogeneous data. The k-nearest neighbor (KNN) join has received a lot of interest in recent years due to its wide range of applications. Processing KNN joins is time-consuming and inefficient due to the quadratic structure of the join method . As the number of applications dealing with vast amounts of data develops, KNN joins get more sophisticated. The authors seek to save money on computer resources by leveraging a large number of threads and multiprocessors. Six popular datasets are used to apply the method and evaluate the sequential and parallel performance of the KNN technique. These datasets are used to compare the sequential and parallel performance of the KNN method. When compared to a matching multi-core solution, the final implementation saves computing resources. It has been optimized to utilize as little RAM as possible, allowing it to manage high-resolution photo data without sacrificing efficiency. The authors will use the technique they presented using Spark Radoop. Our performance research validates the supplied method’s efficacy and scalability.

Comparison of multiclass classification techniques using dry bean dataset

BackgroundThe application of classsification methods through multivariate and machine learning techniques has enormous significance in agricultural sector. It is vital to classify various types of seeds as well as identify the quality of seeds which has a great impact on the production of crops. There is a wide range of genetic variations in dry beans all over the world. Many studies have been conducted previously on various dataset to indentify the sorts of dry beans, however most of them focused on machine learning techniques with binary classification. ObjectiveThe aim of this study is to investigate a reliable classifier which has the lowest noise implications and establish an algorithm for dry bean classification effectively. This paper focuses on outlier removals, oversampling with Adaptive Synthetic (ADASYN) algorithm and finding the best classifier to guarantee the highest possible accuracy. MethodsThe raw dataset for this study was accessed from UCI Machine Learning Repository. The dataset contained grains having 16 features, 12 dimensions, and 4 distinct shapes. For the purpose of eliminating missing values from the dataset, interquartile range (IQR) with python programming was utilized. Eight most popular classifiers were used in this study which are Logistic Regression (LR), Naïve Bayes (NB), k-Nearest Neighbor (KNN), Decision Tree (DT), Random Forest (RF), Extreme Gradient Boosting (XGB), Support Vector Machine (SVM), and Multilayer Perception (MLP) with balanced and imbalanced classes. The authors utilized frequency tables, bar diagrams, boxplots, analysis of variance for descriptive analysis as well as data preprocessing. ResultsThe XGB classifier preferably outperformed than other classifiers with balanced and imbalanced distribution of dry beans within each class. It has acquired accuracy (ACC) 93.0% and 95.4% in imbalanced and balanced classes respectively. In case of balanced dataset, after application of ADASYN algorithm both KNN and RF techniques also performed well regarding the Classification Accuracy (ACC), Sensitivity (SE), Specificity (SP) and Cohen's kappa coefficient (Kappa) etc. The most important attributes for classifying the dry beans were found ShapeFactor2, Minor Axis Length, and ShapeFactor1 along with EquivDiameter, Roundness and ConvexArea. ConclusionsFor classification of dry seeds, the XGB classifier had performed well when the dataset contained both balanced and imbalanced distribution in classes. Also, it is the primary approach of identifying the classes of seeds/beans with balanced or not. If the classes of the target variable are balanced well, then the KNN and RF algorithms may be applied along with XGB technique for more accurate classification.

Application of meta-heuristic feature selection method in low-cost portable device for watermelon classification using signal processing techniques

Recognizing the stage of fruit maturity and thus determining the optimum harvesting time is critical since the competitive market requires high-quality products at a competitive price. Furthermore, in the context of the global water crisis, harvesting watermelons at an inappropriate ripening stage causes water wasting of around 280 kg per kilo of watermelon as a virtual water footprint, highlighting the necessity of criteria for harvesting ripe watermelons by the farmers. This research thus aims to classify the Charleston Gray watermelon type into three categories i.e. unripe, ripe, and overripe using portable acoustic signal processing, data mining methods, and artificial intelligence approaches.Signal processing in the time and frequency domains and Wavelet Transformation were used to extract essential features from acoustic signals of the watermelons. This was followed by the selection of the significant features in classification using a t-test mean comparison. Sample categorization was accomplished using Support Vector Machines and the K-Nearest Neighbor techniques. Considering the effect of three impact positions (the stem side, middle, and flower side), two materials (Polyoxymethylene and metal balls) and two diameters (15 and 10 mm diameter) for striker ball, using Principal Component Analysis and application of various signal processing methods, 3360 models were established. Based on the frequency equipped with the Mahalanobis distance, the K-Nearest Neighbor classifier presented the best result with a 77.3 % classification rate for signals obtained from the small metal ball and stem side impact position. In comparison, experts accurately had categorized 52 % of the samples in total.

Imputing stem frequency distributions using harvester and airborne laser scanner data: a comparison of inventory approaches

Stem frequency distributions provide useful information for pre-harvest planning. We compared four inventory approaches for imputing stem frequency distributions using harvester data as reference data and predictor variables computed from airborne laser scanner (ALS) data. We imputed distributions and stand mean values of stem diameter, tree height, volume, and sawn wood volume using the k-nearest neighbor technique. We compared the inventory approaches: (1) individual tree crown (ITC), semi-ITC, area-based (ABA) and enhanced ABA (EABA). We assessed the accuracies of imputed distributions using a variant of the Reynold’s error index, obtaining the best mean accuracies of 0.13, 0.13, 0.10 and 0.10 for distributions of stem diameter, tree height, volume and sawn wood volume, respectively. Accuracies obtained using the semi-ITC, ABA and EABA inventory approaches were significantly better than accuracies obtained using the ITC approach. The forest attribute, inventory approach, stand size and the laser pulse density had significant effects on the accuracies of imputed frequency distributions, however the ALS delay and percentage of deciduous trees did not. This study highlights the utility of harvester and ALS data for imputing stem frequency distributions in pre-harvest inventories.

3D Face Reconstruction from a Single Image Using a Combined PCA-LPP燤ethod

In this paper, we proposed a combined PCA-LPP algorithm to improve 3D face reconstruction performance. Principal component analysis (PCA) is commonly used to compress images and extract features. One disadvantage of PCA is local feature loss. To address this, various studies have proposed combining a PCA-LPP-based algorithm with a locality preserving projection (LPP). However, the existing PCA-LPP method is unsuitable for 3D face reconstruction because it focuses on data classification and clustering. In the existing PCA-LPP, the adjacency graph, which primarily shows the connection relationships between data, is composed of the e-or k-nearest neighbor techniques. By contrast, in this study, complex and detailed parts, such as wrinkles around the eyes and mouth, can be reconstructed by composing the topology of the 3D face model as an adjacency graph and extracting local features from the connection relationship between the 3D model vertices. Experiments verified the effectiveness of the proposed method. When the proposed method was applied to the 3D face reconstruction evaluation set, a performance improvement of 10% to 20% was observed compared with the existing PCA-based method.

Identifying Potentially Fraudulent Credit Card Transactions with the use of Soft Computing Methods

One of the most pervasive difficulties facing modern finance organisations, corporations, and governments is the alarming rise in fraudulent or dishonest practises around financial transactions, especially those using credit cards. As difficulties go, this is one of the more prevalent ones. To protect the interests of both consumers and businesses, it is crucial to implement a fraud detection system that heavily relies on the use of intelligent fraud detection tactics. Numerous fraud detection methods, strategies, and systems have been reported in the scholarly literature; many of them make use of intelligent strategies like algorithms and frameworks. This project begins out with an exploratory data analysis aimed at identifying potentially fraudulent credit card transactions. The authors next provide some categorization models that are implemented in soft computing by means of ingenious methodologies. For the aim of evaluation, several other classification algorithms such decision trees, random forests, and logistic regressions have been used in addition to the Knearest neighbour (K-NN) technique. The suggested model provides a more accurate, computationally efficient, and lightweight option for detecting credit card fraud.