K-nearest Neighbor Approach Research Articles

Hybrid Model Based on ReliefF Algorithm and K-Nearest Neighbor for Erythemato-Squamous Diseases Forecasting

Machine learning (ML) techniques have been used to solve real-world problems for decades. In the field of medical sciences, these approaches have been found to be useful in the diagnosis and prognosis of a variety of medical disorders. However, when dealing with voluminous, inconsistent, and higher-dimensional data, conventional ML approaches have failed to deliver the expected results. Researchers have suggested hybrid solutions to resolve these problems, which have been found to be more effective than conventional methods because these systems integrate their merits while reducing their drawbacks. In the current research article, hybrid model has been presented by coupling feature optimization with prediction approach. The proposed hybrid model has two stages: the first involves implementing the ReliefF Algorithm for optimum feature selection in erythemato-squamous diseases, and the second involves implementing k-nearest neighbor (KNN) for prediction of those selected optimum features. The experimentation was carried out on bench mark dataset for erythemato-squamous diseases. The presented hybrid model was also assessed with conventional KNN approach based on various metrics such as classification accuracy, kappa coefficient, recall, precision, and f-score.

Regionalization of hydrological models for flow estimation in ungauged catchments in Ireland

Study RegionThe study area consists of 44 catchments across Ireland. Study FocusWe regionalize two hydrological models (GR4J and GR6J) to produce continuous discharge simulations and compare performance in simulating high, median and low flow conditions with other established approaches to prediction in ungauged basins and a simple benchmark of using the median parameter set across all catchments. These include K-nearest neighbor (KNN) and statistical methods for predicting flow quantiles using catchment characteristics. Different objective functions were selected for different parts of flow regime and the success of different methods for regionalizing hydrological model parameters; including multiple linear regression (MLR), non-linear regression (NL) and random forests (RF) were evaluated. New Hydrological insights for the RegionAll regionalization approaches perform well for average flow conditions. The GR4J model regionalized using RF performs best for simulating high flows, though all regionalized models underestimate the median annual flood. GR6J regionalized using RF performs best for low flows. While KNN and statistical approaches that directly leverage physical catchment descriptors provide comparable median performances across catchments, the spread in relative error across our sample is reduced using regionalized hydrological models. Our results highlight that the choice of hydrological model, objective functions for optimization and approach to linking model parameters and physical catchment descriptors significantly influence the success of regionalization for low and high flows.

Single-Cell Transcriptomics of Human Plasma Cells from HLA-Sensitized Patients Reveals Population Heterogeneity and Drug Resistance Adaptations

Kidney transplantation improves the overall survival of end stage renal disease patients but the presence of preformed human leukocyte antigen (HLA)-specific antibodies often precludes finding an acceptable donor. Recent studies demonstrate that bone marrow plasma cells (BMPCs) are a major source of HLA-specific antibodies in highly sensitized transplant candidates. However, therapeutic strategies that target BMPCs and overcome HLA desensitization are only beginning to emerge. Proteasome inhibitor (PI) therapies effectively eliminate BMPCs, reduce HLA-antibody levels and have shown promise as an effective means of enhancing transplantation outcomes. However, PIs are frequently limited by de novo or therapy-induced drug resistance. In addition, BMPCs exhibit variable sensitivity to PI-based therapy. To identify potential mechanisms of drug resistance in BMPCs, we employed single-cell genomic profiling. Single-cell studies provide a powerful tool that allows us to study rare cell populations, such as BMPCs, and intermediate cell states that cannot be resolved at the population level. Here, we have performed single-cell genomic profiling of human BMPCs isolated from highly HLA-sensitized renal transplant candidates prior to and after desensitization therapy with the irreversible PI carfilzomib. HLA-sensitized transplant candidates received CFZ (20, 27, 36 mg/m2) with BM aspirates performed prior to and after CFZ therapy. CD138+BMPCs were isolated to >98% purity and individual cells analyzed using the recently developed method, Iterative Clustering and Guide-gene Selection (ICGS), which utilizes pair-wise correlation of dynamically expressed genes and iterative clustering with pattern-specific guide genes to delineate coherent gene-expression patterns. Strongly and differentially expressed gene sets were clustered and evaluated using gene enrichment, prior knowledge-based networks and ICGS. Each cluster corresponds to one or a group of closely related cell types that represent spatial or temporal distinct states. Single-cell profiling by the drop-seq method analyzed mRNA expression in 2,000 individual BMPCs isolated from the same HLA-sensitized patient prior to and after CFZ therapy. We used CellHarmony, a new bioinformatic approach that allows us to determine the frequency of all CFZ-treated cells in direct reference to the untreated cells and any altered gene expression profiles. In our first comparison of these cells, eight cell clusters were defined (Fig. 1A). Single-cell analysis of BMPCs obtained post-CFZ therapy indicated qualitative and quantitative changes in gene expression as well differences in the proportionality of existing cell clusters. To determine the frequency and expression of these genes in the CFZ-treated samples, we then performed classification of those samples using a k-nearest neighbor approach using all untreated cells as the reference. For example, protein catabolism, nonsense-mediated decay and integrin signaling were increased (clusters 1, 5 and 6) while clusters enriched in IL-12 signaling, interferon signaling and protein translation genes (2, 3 and 7) were reduced. Importantly, the state associated with the mitotic cell cycle was significantly upregulated in BMPCs that survived CFZ. BMPCs were analyzed from additional patients using a comparable approach identified common gene expression sets in pre-treatment samples as well as recurrent differences in post-treatment samples (Fig. 1B).Conclusions: The results presented here demonstrate the power and promise of single-cell genomics in uncovering functional diversity between cells and in deciphering how cell states respond to therapeutic challenge. Our results indicate a shift toward highly mitotic states in BMPCs that survive CFZ therapy. Importantly, the single cell results also provide internal validation of our prior bulk RNA-seq studies on BMPCs. Our results suggest the feasibility of pathway-based secondary targeting of the recovering cell populations to enhance response. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Using Individualised HDI Measures for Predicting Educational Performance of Young Students—A Swedish Case Study

HDI is a frequently used quantitative index of human potential and welfare, developed as a comprehensive measure for the cross-sectional and temporal comparison of socioeconomic performance. The HDI is a standardised quantitative estimation of welfare comprising indicators of health, knowledge and standard of living, enabling assessment over countries, regions or time periods, in case of limited data access. The index highlights critical conditions for equity and socioeconomic development outside the group of stakeholders and researchers. The HDI provides a learning potential that may be harnessed to enhance insights into the magnitude of human potential at super-local levels. In this paper we design, implement and test the validity of a super-local variant of HDI in the context of pedagogical performance of young pupils. We compare whether HDI is a good predictor for school grades among all ninth-grade students in Sweden during the year 2014. Our results show that a super-local HDI index is performing equal to or better than the one related to standard measures of human potential, while the index can be generated on individual levels using k-nearest neighbour approaches during the index creation process.

Two-level K-nearest neighbors approach for invasive plants detection and classification

Invasive plants (IPs) are considered major threat to biodiversity and ecosystems due to their potential to compete for resources with other plants or crops. Timely and effective detection is important to prevent and control their growth. Recent advancements in unmanned aerial systems (UAS) have brought an effective, inexpensive, and non-invasive approach to monitoring plants. The use of UAS is promising in the accurate and rapid identification of IPs in the field. However, challenges exist when using UAS for IP detection as they generate a huge number of images that must be interpreted by researchers manually. This time-consuming and error-prone process significantly limits the applicability of UAS in a variety of environmental and agriculture monitoring applications. To reduce processing time and increase detection accuracy, we develop an automatic and intelligent detection approach to identify IPs with high levels of efficiency and accuracy from images of the monitoring areas captured by UAS. First, the original field images are segmented into patch images using the simple linear iterative clustering (SLIC) superpixels algorithm. Then a novel two-level K-nearest neighbors (TLKNN) algorithm is proposed to classify the patch images into different IP categories. In it, a deep convolutional neural network (CNN) is used to extract the features from the patch image and a kernel map function is utilized to obtain the kernel features. Two feature sets are fed to train two K-nearest neighbor (KNN) models. Finally, the final classification is determined by the selection results by KNN models. In the experiments, two public UAV image sets are employed to train and verify the proposed method. Several metrics such as accuracy, recall, precision, and F1-score, are used to evaluate the performance. The proposed TLKNN is compared with two newly proposed convolutional neural network (CNN) methods on the same datasets. Experimental results indicate the proposed TLKNN algorithm achieves better classification performance and reaches the highest accuracy, recall rate, precision rate, and F1-score. The improved performance not only means that IPs can be detected more accurately and automatically, but also that there are environmental benefits in developing an intelligent strategy for IP prevention and control.

Spatial and temporal variability of water stress risk in the Kebir Rhumel Basin, Algeria

This paper presents a methodological framework to conduct a water stress assessment of water resources through water stress index. A modified version of the water stress index was developed and applied for cereal grown in the Kebir Rhumel Basin, Algeria. The water stress index defined as Crop Stress Index (CSI) and derived by accumulating differences in water supply and demand over time to assess the maximum cumulative deficit. The forecasting crop water stress index for cereals was developed here as a way of understanding and hedging risk. The CSI was forecasted using the appropriate climate predictors as predictors in a non-parametric k-nearest neighbors (k-NN) approach that issues probabilistic forecasts. The accuracy of the CSI forecasts in the years 2002–2013 was judged based directional concordance and contingency metrics such as hit, miss and false alarm rate. The results showed that our forecasts were correct 8 out of 12 times, with three misses and just one false alarm. These results confirmed that this is a useful approach in investigating irrigation needs. Finally, it was found that this water stress index is a useful tool to estimate water storage and irrigation requirements in the interest of agricultural water resources.

419 STOP-Bang Score and History of Radiation Predicts Risk of Obstructive Sleep Apnea in Cancer Patients: A Machine Learning Study

Abstract Introduction Cancer patients are at an increased risk of moderate-to-severe obstructive sleep apnea (OSA). The STOP-Bang score is a commonly used screening questionnaire to assess risk of OSA in the general population. We hypothesize that cancer-relevant features, like radiation therapy (RT), may be used to determine the risk of OSA in cancer patients. Machine learning (ML) with non-parametric regression is applied to increase the prediction accuracy of OSA risk. Methods Ten features namely STOP-Bang score, history of RT to the head/neck/thorax, cancer type, cancer stage, metastasis, hypertension, diabetes, asthma, COPD, and chronic kidney disease were extracted from a database of cancer patients with a sleep study. The ML technique, K-Nearest-Neighbor (KNN), with a range of k values (5 to 20), was chosen because, unlike Logistic Regression (LR), KNN is not presumptive of data distribution and mapping function, and supports non-linear relationships among features. A correlation heatmap was computed to identify features having high correlation with OSA. Principal Component Analysis (PCA) was performed on the correlated features and then KNN was applied on the components to predict the risk of OSA. Receiver Operating Characteristic (ROC) - Area Under Curve (AUC) and Precision-Recall curves were computed to compare and validate performance for different test sets and majority class scenarios. Results In our cohort of 174 cancer patients, the accuracy in determining OSA among cancer patients using STOP-Bang score was 82.3% (LR) and 90.69% (KNN) but reduced to 89.9% in KNN using all 10 features mentioned above. PCA + KNN application using STOP-Bang score and RT as features, increased prediction accuracy to 94.1%. We validated our ML approach using a separate cohort of 20 cancer patients; the accuracies in OSA prediction were 85.57% (LR), 91.1% (KNN), and 92.8% (PCA + KNN). Conclusion STOP-Bang score and history of RT can be useful to predict risk of OSA in cancer patients with the PCA + KNN approach. This ML technique can refine screening tools to improve prediction accuracy of OSA in cancer patients. Larger studies investigating additional features using ML may improve OSA screening accuracy in various populations Support (if any):

SME 4.0: Machine learning framework for real-time machine health monitoring system

Over the past ten years, manufacturers and consumers have become increasingly interested in the applications of smart, sustainable, and autonomous systems in the industry and in everyday life. Due to the recent industrial revolution (Industry 4.0), most of the existing Small and Medium-sized Enterprises (SMEs) also want to adapt their work environment into the smart system by the applications of these technologies such as the Industrial Internet of Things (IIoT), Artificial Intelligence (AI) techniques, Machine Learning Algorithm (MLA), Internet Communication Technology (ICT), and Cyber-Physical System (CPS). Because they are very much interested in maximizing productivity, machine availability, reliability, and customer satisfaction in this competitive industrial world. This research study particularly focuses on the Predictive Maintenance (PdM) activity of critical machines and their components in the SME based on the maintenance history dataset through the application of the supervised machine learning algorithm such as Logistics Regression (LR) and K-Means (K-Nearest Neighbor) approaches. In accordance, the real-time case study is presented in SMEs in the southern region of Tamil Nadu, India with two-phase activities. Initially, the optimal failure rate of the machines is predicted by the utilization of LR trained models. Then trigger the man-machine communication and suitable decision-making process of service and maintenance activity through the application of the K-Mean approaches. The main objective of this research study is to organize the smart PdM activity of the smart factory systems in SMEs with the application of MLA based on the real-time maintenance history dataset.

Rumor Situation Discrimination Based on Empirical Mode Decomposition Correlation Dimension

To effectively identify network rumors and block their spread, this paper uses fractal theory to analyze a network rumor spreading situation time series, reveal its inner regularity, extract features, and establish a network rumor recognition model. The model is based on an empirical mode decomposition (EMD) correlation dimension and K-nearest neighbor (KNN) approach. Firstly, a partition function is used to determine if the time series of the rumor spreading situation is a uniform fractal process. Secondly, the rumor spreading situation is subjected to EMD to obtain a series of intrinsic mode functions (IMFs), construct the IMF1–IMF6 components containing effective feature information as the principal components, and reconstruct the phase space of the principal components, respectively. Finally, the correlation dimensions of the principal components IMF1–IMF6 as obtained by the Grassberger-Procaccia algorithm are used as feature parameters and are imported into the KNN model for rumor recognition. The experimental results show that the correlation dimension of a spreading situation can better reflect the characteristic information; as combined with the KNN model for identifying rumors, the recognition rate reaches 87.5%. This result verifies the effectiveness of fractal theory in network rumors recognition, expands the thinking for the research of rumors recognition, and provides theoretical support for rumor governance.

Fuzzy decision ontology for melanoma diagnosis using KNN classifier

Melanoma is the most dangerous type of skin cancer when discovered in an advanced stage. Early detection of melanoma improves survival. Several Computer -Aided Diagnosis (CAD) systems are currently developed to speed up early diagnosis. Recently, ontology is widely adapted for describing and diagnosing a disease. For melanoma detection, the ontology reasoning of dermatologists is based on expert rules, such as ABCD rule. Accordingly, dermatologists classify skin lesions in three classes: melanoma, benign, and recommended follow-up class. In this paper, we propose a CAD system based on an ontology for melanoma diagnosis by giving the probability of being melanoma. We first present our ontology focusing on its main concepts involved in ABCD rule: Asymmetry, Border, Color and Differential structures. Accordingly, the Bag-of-Words, modeling these concepts, are generated from extracted features of skin lesion images. An important step in ontology is to define rules relating the different concepts. In our case, these rules allow the fuzzy decision to classify lesion in melanoma, benign or recommended follow-up class with a malignancy probability. Considering the similarity of melanoma cases, the K-Nearest Neighbors approach is applied to make the final decision in case of a recommended follow-up class. Experimental validation on two public datasets of 206 lesion images shows that our approach presents an efficient method of analysis and can be more appropriate for lesion severity classification. It yields a sensitivity of (96%) and an accuracy of (92%), surpassing existing recent approaches on melanoma diagnosis.

Computing Expectiles Using k-Nearest Neighbours Approach

Expectiles have gained considerable attention in recent years due to wide applications in many areas. In this study, the k-nearest neighbours approach, together with the asymmetric least squares loss function, called ex-kNN, is proposed for computing expectiles. Firstly, the effect of various distance measures on ex-kNN in terms of test error and computational time is evaluated. It is found that Canberra, Lorentzian, and Soergel distance measures lead to minimum test error, whereas Euclidean, Canberra, and Average of (L1,L∞) lead to a low computational cost. Secondly, the performance of ex-kNN is compared with existing packages er-boost and ex-svm for computing expectiles that are based on nine real life examples. Depending on the nature of data, the ex-kNN showed two to 10 times better performance than er-boost and comparable performance with ex-svm regarding test error. Computationally, the ex-kNN is found two to five times faster than ex-svm and much faster than er-boost, particularly, in the case of high dimensional data.

An Efficient Class-Based Data Clustering Through K-Means And Knn Approach

In this paper an efficient class-based data clustering through k-means and KNN approach were applied. It has been applied for proper data grouping and efficient classification. In this approach three dataset have been considered for the experimentation. Data preprocessing has been performed for the removal of unmatched or empty entry. Then weight assignment and normalization has been performed. K-means and k-nearest neighbor (KNN) have been applied on the dataset for the data grouping and classification purpose. Different splitting variations have been considered with higher variance. Data selection is completely random. The results obtained shows the strength of our approach though classification and class-based clustering.

Design and Additive Manufacturing of Porous Sound Absorbers-A Machine-Learning Approach.

Additive manufacturing (AM), widely known as 3D-printing, builds parts by adding material in a layer-by-layer process. This tool-less procedure enables the manufacturing of porous sound absorbers with defined geometric features, however, the connection of the acoustic behavior and the material’s micro-scale structure is only known for special cases. To bridge this gap, the work presented here employs machine-learning techniques that compute acoustic material parameters (Biot parameters) from the material’s micro-scale geometry. For this purpose, a set of test specimens is used that have been developed in earlier studies. The test specimens resemble generic absorbers by a regular lattice structure based on a bar design and allow a variety of parameter variations, such as bar width, or bar height. A set of 50 test specimens is manufactured by material extrusion (MEX) with a nozzle diameter of mm and a targeted under extrusion to represent finer structures. For the training of the machine learning models, the Biot parameters are inversely identified from the manufactured specimen. Therefore, laboratory measurements of the flow resistivity and absorption coefficient are used. The resulting data is used for training two different machine learning models, an artificial neural network and a k-nearest neighbor approach. It can be shown that both models are able to predict the Biot parameters from the specimen’s micro-scale with reasonable accuracy. Moreover, the detour via the Biot parameters allows the application of the process for application cases that lie beyond the scope of the initial database, for example, the material behavior for other sound fields or frequency ranges can be predicted. This makes the process particularly useful for material design and takes a step forward in the direction of tailoring materials specific to their application.

A reliable KNN filling approach for incomplete interval-valued data

Interval-valued data (IVD) is a kind of data where each feature is an interval, and embeds the uncertainty and variability information. However, the missing values (lower or upper bound, or both of them are missed) may occur in the process of data acquisition and transmission, which may lead to obstacles for data processing. To obtain good results, it is important for IVD to process (often ignore or fill) the missing values. A dataset including missing values is named as incomplete interval-valued (IIV) set here. Some ignoring and filling methods for numeric or symbolic data have been proposed, but they cannot be applied for IIV datasets directly. In this work, a reliable k-nearest neighbor approach (RKNN) for incomplete interval-valued data (IIVD) is proposed. A combining rule to determine whether a datum including missing values should be ignored or filled is designed. Those samples with the missing value for each feature will be ignored directly. It is different from existing ignoring methods that need to set the percentage of missing entries. For the rest of missing samples, they will be filled according to their K complete nearest neighbors, which can ensure the filled value more reliable. In so doing, RKNN can exclude a small number of missing samples that may increase uncertainty, and avoid the repetition of the filled values (like median or a fixed constant). The experiment results on 12 synthetic datasets and 4 real-world datasets demonstrate that the proposed method can process the incomplete interval-valued data effectively, and obtain a good classification performance simultaneously.

Prediction of Succinic Acid Extraction Efficiency in the Emulsion Liquid Membrane by using Machine Learning Techniques

This research aims to predict succinic acid concentration in the external phase during the emulsion liquid membrane process by using artificial neural networks along with a popular alternative method: k-nearest neighbor technique. The solute concentration values can be predicted by the proposed method without performing a great number of emulsion liquid membrane experiments. Several computer simulations were performed to demonstrate the success of the system. Simulation results showed that the estimated solute concentration values are very close to the achieved experimental results. The optimal conditions for emulsion liquid membrane were found to be: solvent kerosene, TOPO concentration (1%w/w), Amberlite LA-2 concentration (4%w/w), surfactant concentration (5%w/w), Na2CO3 concentration (5%w/v), modifier (decanol) concentration (2%w/w), mixing speed 300 rpm. The average accuracy percentages achieved by artificial neural network and k-nearest neighbor approaches were 88.75and#177;1.94% and 90.2and#177;1.2%, respectively.

Prediction of Succinic Acid Extraction Efficiency in the Emulsion Liquid Membrane by using Machine Learning Techniques

This research aims to predict succinic acid concentration in the external phase during the emulsion liquid membrane process by using artificial neural networks along with a popular alternative method: k-nearest neighbor technique. The solute concentration values can be predicted by the proposed method without performing a great number of emulsion liquid membrane experiments. Several computer simulations were performed to demonstrate the success of the system. Simulation results showed that the estimated solute concentration values are very close to the achieved experimental results. The optimal conditions for emulsion liquid membrane were found to be: solvent kerosene, TOPO concentration (1%w/w), Amberlite LA-2 concentration (4%w/w), surfactant concentration (5%w/w), Na2CO3 concentration (5%w/v), modifier (decanol) concentration (2%w/w), mixing speed 300 rpm. The average accuracy percentages achieved by artificial neural network and k-nearest neighbor approaches were 88.75and#177;1.94% and 90.2and#177;1.2%, respectively.

High-Speed and Accurate Meat Composition Imaging by Mechanically-Flexible Electrical Impedance Tomography With k-Nearest Neighbor and Fuzzy k-Means Machine Learning Approaches

High-speed and accurate meat composition imaging method has been proposed based on mechanically-flexible electrical impedance tomography (mech-f-EIT) with k-nearest neighbor and fuzzy k-means machine learning approaches. This proposed method has four stages which are 1) estimation of meat boundary shape ∂Ω by mech-f-EIT for base data, 2) approximation of Jacobian matrix J* by k-nearest neighbor (k-NN) algorithm under ∂Ω for high speed, 3) clustering of meat composition <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</sup> σ (fat k = 1, lean k = 2, bone k = 3) by fuzzy k-means algorithm based on the reconstructed meat conductivity distribution σ for high accuracy, and 4) edge detection of meat composition <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</sup> Ω by Canny algorithm for sharp edge. This method is qualitatively evaluated by using two agar phantoms, a cow's lower leg and three lamb's lower legs. As the results, mech-f-EIT estimates ∂Ω with total mean boundary error 〈ẽ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">b</sub> 〉 = 4.81 %. This method achieves high-speed approximation of J* with total mean speed-up performance 〈s̃p〉 = 4.51 times as compared with the computation time of standard J; nonetheless, total mean cross correlation between J* and J is accurate 〈c̃ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">c</sub> 〉 = 0.92. Moreover, this method clusters the <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</sup> σ with total mean area error 〈ẽ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a</sub> 〉 = 4.49 %. Furthermore, this imaging method detects sharply the meat composition edges <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</sup> Ω between fat and lean (k = 1 - 2) and between lean and bone (k = 2 - 3) with total mean edge error 〈ẽ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</sub> 〉 =6.90 %.

Multistage Ensemble Learning Model With Weighted Voting and Genetic Algorithm Optimization Strategy for Detecting Chronic Obstructive Pulmonary Disease

Chronic Obstructive Pulmonary Disease (COPD) is a life-threatening lung ailment and a significant cause of morbidity and fatality globally. The early detection of COPD can provide timely proper medication and reduce the mortality rate. To obtain proper treatment and lessen the death rate, this study proposes a novel ensemble model: the Multistage Ensemble model (MSEN) with an optimized weighted voting technique to detect COPD early and help clinicians provide proper and timely medication. In this study, there are two pools of classifiers created in which four classifiers are placed in each pool. These two pools of classifiers are employed to form two weighted ensemble models based on a weighted voting strategy. This study combines those generated ensemble models using a weighted voting technique to form an MSEN model. The genetic algorithm is utilized to optimize the hyperparameters of each classifier in each pool. The weights of two generated ensemble models and each classifier are optimized using the grid search technique. This study employs the K-Nearest Neighbors approach to fill in the missing values, isolation forest to remove the outliers, and the LightGBM with Recursive Feature Elimination for feature selection. An evaluation of the suggested MSEN model is conducted on a real-world Exasens dataset to validate the suggested model's effectiveness which exhibits that the proposed model obtains better performance to detect COPD and provides superior performance than other machine learning models and existing benchmark techniques.

GeneticKNN: a weighted KNN approach supported by genetic algorithm for photometric redshift estimation of quasars

We combine K-nearest neighbors (KNN) with a genetic algorithm (GA) for photometric redshift estimation of quasars, short for GeneticKNN, which is a weighted KNN approach supported by a GA. This approach has two improvements compared to KNN: one is the feature weighted by GA; the other is that the predicted redshift is not the redshift average of K neighbors but the weighted average of median and mean of redshifts for K neighbors, i.e. p × z median + (1 – p) × z mean. Based on the SDSS and SDSS-WISE quasar samples, we explore the performance of GeneticKNN for photometric redshift estimation, comparing with the other six traditional machine learning methods, i.e. the least absolute shrinkage and selection operator (LASSO), support vector regression (SVR), multi-layer perceptrons (MLP), XGBoost, KNN and random forest. KNN and random forest show their superiority. Considering the easy implementation of KNN, we make improvement on KNN as GeneticKNN and apply GeneticKNN on photometric redshift estimation of quasars. Finally the performance of GeneticKNN is better than that of LASSO, SVR, MLP, XGBoost, KNN and random forest for all cases. Moreover the accuracy is better with the additional WISE magnitudes for the same method.

Exchange Market Liquidity Prediction with the K-Nearest Neighbor Approach: Crypto vs. Fiat Currencies

In this paper, we compare the predictions on the market liquidity in crypto and fiat currencies between two traditional time series methods, the autoregressive moving average (ARMA) and the generalized autoregressive conditional heteroskedasticity (GARCH), and the machine learning algorithm called the k-nearest neighbor (KNN) approach. We measure market liquidity as the log rates of bid-ask spreads in a sample of three cryptocurrencies (Bitcoin, Ethereum, and Ripple) and 16 major fiat currencies from 9 February 2018 to 8 February 2019. We find that the KNN approach is better suited for capturing the market liquidity in a cryptocurrency in the short-term than the ARMA and GARCH models maybe due to the complexity of the microstructure of the market. Considering traditional time series models, we find that ARMA models perform well when estimating the liquidity of fiat currencies in developed markets, whereas GARCH models do the same for fiat currencies in emerging markets. Nevertheless, our results show that the KNN approach can better predict the log rates of the bid-ask spreads of crypto and fiat currencies than ARMA and GARCH models.