K-nearest Neighbor Method Research Articles

DEVELOPMENT OF A CREDIT SCORING MODEL USING MACHINE LEARNING FOR COMMERCIAL BANKS IN VIETNAM

In an increasingly competitive financial market, developing an accurate and efficient credit scoring model is crucial for banks to make informed credit decisions, and to help customers obtain credit products that match their financial capabilities. This study aims to develop a credit scoring model for individual customers of Vietnamese banks using machine learning techniques. This article will use a set of financial and non-financial indicators as inputs for various machine learning models such as Logistic Regression, K-nearest Neighbor, Decision Tree, Random Forest, LightGBM, and Support Vector Method. The obtained results from these models will be carefully compared and evaluated to ultimately select the best credit scoring model for the bank. This study leverages Google Colab, a cloud-based platform, for comprehensive data analysis and model development, ensuring a robust and data-driven approach. This study hopes to provide a useful solution for Vietnamese banks in managing credit risk and improving their business performance.

Objects Diversity and its Impact on Classification Quality in Dispersed Data Environments

This paper studies a classification model dedicated to dispersed data, employing the k-nearest neighbors method as a base classifier and a Radial Basis Function (RBF) network as a fusion method. Focusing on classifying objects described by different attributes, the study systematically reduces common objects in local tables to assess the robustness of the model. Surprisingly, the proposed approach shows resilience in reducing common objects without significantly affecting key metrics such as F-measure, balanced accuracy and overall accuracy. Moreover, the studied model performs better on balanced data. This research contributes valuable insights into dispersed data classification, demonstrating the model’s effectiveness in handling diverse objects and attributes. The findings have implications for fields reliant on dispersed data storage, such as healthcare, banking, and surveillance, showcasing the model’s potential for real-world applications.

STRUCTURE-ACTIVITY RELATIONSHIP OF PSYCHOACTIVE CANNABINOIDS: A CHEMOMETRIC EXPERIMENT FOR UNDERGRADUATE AND POSTGRADUATE STUDENTS IN CHEMICAL EDUCATION

Undergraduate and postgraduate Chemistry Education students had to develop the experiment in four (4) stages. Twenty-two (22) cannabinoids (Δ9-THC and derivatives) with different degrees of psychoactivity, previously scrutinized by our research group with the approaches: molecular electrostatic potential (MEP) and interaction with the CB1 receptor (CARDOSO-FILHO et al., 2004), are investigated through Chemometrics. Exploratory Analysis: Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA). E Classification Methods: K-Nearest Neighbor (KNN) method, Soft Independent Modeling Class Analogy (SIMCA) method, and Stepwise Discriminant Analysis (SDA) were employed to reduce the dimensionality of the data matrix and investigate which descriptors are responsible for the classification between the most active cannabinoids (mac), pKi ≥ 6.72, and the least active cannabinoids (lac), pKi < 6.72, according to the hypothesis previously reported (Cardoso-Filho et al., 2024). The investigation with PCA, HCA, KNN, SIMCA, and SDA showed that the descriptors LUMO+1 energy, Geary autocorrelation of lag 6 weighted by van der Waals volumes (GATS6V), hydration energy (EH), atomic charge on the atom of C4 (qC4), and molecular representation of structure based on the electron diffraction, code of signal 3, unweighted (MOR03u) are responsible for separating cannabinoids according to their degrees of psychoactivity. The insights accumulated and the models built in the research – PCA chemometric model, HCA chemometric model, KNN chemometric model, SIMCA chemometric model, and SDA chemometric model – will be able to support the design of new potentially psychoactive Δ9-THC derivatives.

Analisis Sentimen Terhadap Sistem Informasi Akademik Mahasiswa pada Aplikasi Edlink dengan Metode K-Nearest Neighbor

The Sevima Edlink application is an academic information system that is widely used by educational institutions in Indonesia to manage student academic data and information. Although this application has various useful features, its successful implementation also depends greatly on user satisfaction and acceptance. Therefore, it is important to analyze user sentiment towards these applications to identify existing strengths and weaknesses. This research aims to analyze user sentiment towards the Sevima Edlink application using the K-Nearest Neighbor (K-NN) method. The K-NN method was chosen because of its simplicity and effective ability to classify sentiment data. The data used in this research are reviews from application users collected from various sources. The results of this research used the K-NN method, namely an accuracy value of 94.47%. So it can be said that the K-KNN algorithm can classify data well and correctly.

A Hybrid Algorithm of Ship Trajiectories based on Random Forest and K-Nearest Neighbor

Abstract The technology for predicting ship trajectories is crucial in enhancing maritime efficiency, and ensuring maritime safety and environmental protection. This paper introduces a hybrid algorithm that combines Random Forest with the K-Nearest Neighbor method, and various machine learning techniques, such as Extreme Gradient Boosting, Support Vector Regression, and Random Forests, are explored and compared for ship trajectory prediction. On a public dataset, a Bayesian optimization-based training method was used and the model was evaluated by multiple evaluation metrics. The results show that a hybrid algorithm combining Random Forest and K-Nearest Neighbor performs well in predicting ship trajectories and significantly improves prediction accuracy and stability.

Machine learning enabled compact flexible full ground UWB antenna for wearable applications

Abstract This work introduces a novel compact ultra-wideband (UWB) antenna designed for wearable applications, employing a bioinspired structure and machine learning (ML) techniques to achieve exceptional performance in the 3.10–10.42 GHz range. The antenna is fabricated by positioning conductive fabric on a polydimethylsiloxane polymer of 1 mm thickness to augment high flexibility and durability. Additionally, it pioneers integrating a complete ground plane to mitigate back radiation toward the human body, presenting a compact (35.5 × 30.5 × 1 mm3) UWB antenna design compliant with IEEE 802.15.6 standards. The design methodology includes using bandwidth enhancement techniques such as chamfering edges, slots, and adding stubs in the feed, along with applying ML to optimize the antenna’s dimensions for desired return loss characteristics. The proposed antenna demonstrates exceptional resilience to human body loading and physical deformation. The simulation and measurement results have good agreement. The K-nearest neighbour method beat the other ML algorithms maximum accuracy of 99.62% to predict the S11. According to the author’s best knowledge, this is the first compact UWB antenna with full ground specified by IEEE.802.15.6 with ML reported.

EOG Region Detection using Fuzzy k-NN for Virtual Reality Applications

Virtual Reality (VR) systems have become widespread for a decade with the mass production of VR headsets. Advancement in the VR industry benefits both biomedical and computer gaming fields to create better Human-Computer Interface (HCI) applications. In this study, Electrooculogram (EOG) signals are studied on a calibrated A4 paper to simulate reading and tracking eye movement in different regions for VR user interface applications. For that reason, eye activity features from EOG are used to identify relative 2D spatial coordinates and classified with the fuzzy k-Nearest Neighbor (fuzzy k-NN) method. Within the experimental setup, different behaviors such as blinking and depth focus change signals are recorded with constant depth regional borders are analyzed on an A4 paper with reading eye movement recordings. In experimental results, fuzzy k-NN classification results are obtained from observed regional eye movement. The study shows that the fuzzy k-NN method to detect regions at a reading distance is feasible for user interface applications in VR. So, by setting rendering focus at the detected regional area, eye strain can be reduced during prolonged VR sessions especially when reading and/or on user interfaces.

Rapid determination of the geographical origin of kimchi by Fourier transform near-infrared spectroscopy coupled with chemometric techniques.

Determining the geographical origin of kimchi holds significance because of the considerable variation in quality and price among kimchi products from different regions. This study explored the feasibility of employing Fourier transform near-infrared spectroscopy in conjunction with supervised chemometric techniques to differentiate domestic and imported kimchi products. A total of 30 domestic and 30 imported kimchi products were used to build datasets. Three categories of preprocessing methods such as scattering correction (multiplicative signal correction and standard normal variate), spectral derivatives (the first derivative and the second derivative), and data smoothing (Savitzky-Golay filtering and Norris derivative filtering) were used. K-nearest neighbors, support vector machine, random forest, and partial least squares-discriminant analysis were employed. By appropriately preprocessing spectral data, these four methods successfully distinguished between the two sample groups based on their origin. Notably, the k-nearest neighbors method exhibited exceptional performance, accurately classifying the sample groups irrespective of the preprocessing method employed and swiftly achieving this classification. In comparison, classification and regression tree as well as naïve Bayes methods were outperformed by the aforementioned four classification techniques. Particularly, the efficiency and accuracy of the k-nearest neighbors method make it the most recommended chemometric tool for determining the geographical origins of kimchi.

Application of the K-Nearest Neighbor Method for Classification of Hypertension Diseases (Case Study: Stabat Health Center)

Globally, the WHO (World Health Organization) estimates that non-communicable diseases cause about 60% of deaths and 43% of diseases worldwide. Hypertension is a disease that occurs due to an increase in blood pressure in humans. It is difficult to know if a person has hypertension, without measuring the patient's blood pressure. According to the American Heart Association (AHA), the number of Americans over the age of 20 suffering from hypertension has reached 74.5 million, but nearly 90-95% of cases have no known cause. It is estimated that about 80% of the increase in hypertension cases will occur mainly in developing countries by 2025, from 639 million cases in 2000. This number is expected to increase to 1.15 billion cases in 2023. This study uses a quantitative approach with experimental methods to test the application of K-Nearest Neighbor (KNN) in the classification of hypertension diseases at the Stabat Health Center. The description of the results obtained is to make the right decision regarding when and how to treat the disease to prevent the worst possibility for patients by classifying the severity of hypertension both in normal circumstances, prehypertension, stage 1 hypertension, and stage 2 hypertension. The results of the trial show that the KNN model is able to provide accurate predictions based on patient history data available at Stabat Health Center.

COMPARISON OF NAÏVE BAYES AND K-NEAREST NEIGHBOR MODELS FOR IDENTIFYING THE HIGHEST PREVALENCE OF STUNTING CASES IN EAST JAVA

Indonesia will experience a demographic bonus in 2030, where the productive age group will dominate the population and become a buffer for the economy. However, this potential is in vain if human resources experience stunting. According to WHO (2015), stunting is a disorder of child growth and development due to chronic malnutrition and repeated infections, characterized by below-standard length or height. Based on the background of the problem, the author wants to compare the prediction of the prevalence of the highest stunting cases in East Java using the Naive Bayes method and the K-Nearest Neighbor method. The stages carried out in this study are data collection, initial data processing, advanced data processing using the Naïve Bayes Method and K-Nearest Neighbor, and comparative analysis. The results of the implementation of the Naïve Bayes and K-Nearest Neighbor methods are in the form of stunting prevalence prediction charts with variables that affect LBW and TTD. The results of simulations conducted in 6 regions, the Naive Bayes method gets the highest accuracy value of 83.33% in simulation one and 66.67%. The smallest RMSE value is 0.382 simulation 1 and 0.469 simulation 2. This shows that the Naive Bayes method can predict well.

The Impact of Sample Quantity, Traceability Scale, and Shelf Life on the Determination of the Near-Infrared Origin Traceability of Mung Beans.

This study aims to address the gap in understanding of the impact of the sample quantity, traceability range, and shelf life on the accuracy of mung bean origin traceability models based on near-infrared spectroscopy. Mung beans from Baicheng City, Jilin Province, Dorbod Mongol Autonomous, Tailai County, Heilongjiang Province, and Sishui County, Shandong Province, China, were used. Through near-infrared spectral acquisition (12,000-4000 cm-1) and preprocessing (Standardization, Savitzky-Golay, Standard Normal Variate, and Multiplicative Scatter Correction) of the mung bean samples, the total cumulative variance contribution rate of the first three principal components was determined to be 98.16% by using principal component analysis, and the overall discriminatory correctness of its four origins combined with the K-nearest neighbor method was 98.67%. We further investigated how varying sample quantities, traceability ranges, and shelf lives influenced the discrimination accuracy. Our results indicated a 4% increase in the overall correct discrimination rate. Specifically, larger traceability ranges (Tailai-Sishui) improved the accuracy by over 2%, and multiple shelf lives (90-180-270-360 d) enhanced the accuracy by 7.85%. These findings underscore the critical role of sample quantity and diversity in traceability studies, suggesting that broader traceability ranges and comprehensive sample collections across different shelf lives can significantly improve the accuracy of origin discrimination models.

Comparison of clustering methods in determining gold mineralization anomalies in Janja area, Iran

Janja region is located 30 kilometers northwest of Sefidabeh city in Sistan and Baluchistan province, Iran. Considering the evidence of gold mineralization, it is necessary to identify the promising mineral areas in this region. Clustering is one of the important methods in data mining science, which is used to analyze and examine a significant amount of data with various characteristics. The purpose of this research is to investigate the ability of clustering methods to determine gold anomaly areas in the studied area and compare their performance with each other. Hence, K-Means and K-Medoids clustering algorithms have been used to investigate the behavior of gold and copper elements towards each other. For this purpose, first, the number of clusters was changed from k=3 to k=11 and the optimal number of clusters was obtained using Davis-Bouldin, Silhouette and Calinski-Harabasz criteria. According to the obtained results, it can be said that both methods for estimating the gold grade are suitable and effective methods due to the high correlation coefficient. Also, the widely used K-Nearest Neighbor (KNN) method was used under the grade values of gold, silver, copper, lead and zinc in order to simulate and determine the promising areas of gold mineralization. The results show that the KNN method has been effective for finding gold anomalies with proper performance.

Sensor-Aware Data Imputation for Time-Series Machine Learning on Low-Power Wearable Devices

Wearable devices that have low-power sensors, processors, and communication capabilities are gaining wide adoption in several health applications. The machine learning algorithms on these devices assume that data from all sensors are available during runtime. However, data from one or more sensors may be unavailable due to energy or communication challenges. This loss of sensor data can result in accuracy degradation of the application. Prior approaches to handle missing data, such as generative models or training multiple classifiers for each combination of missing sensors are not suitable for low-energy wearable devices due to their high overhead at runtime. In contrast to prior approaches, we present an energy-efficient approach, referred to as Sensor-Aware iMputation (SAM), to accurately impute missing data at runtime and recover application accuracy. SAM first uses unsupervised clustering to obtain clusters of similar sensor data patterns. Next, it learns inter-relationship between clusters to obtain imputation patterns for each combination of clusters using a principled sensor-aware search algorithm. Using sensor data for clustering before choosing imputation patterns ensures that the imputation is aware of sensor data observations. Experiments on seven diverse wearable sensor based time-series datasets demonstrate that SAM is able to maintain accuracy within 5% of the baseline with no missing data when one sensor is missing. We also compare SAM against generative adversarial imputation networks (GAIN), transformers, and k-nearest neighbor methods. Results show that SAM outperforms all three approaches on average by more than 25% when two sensors are missing with negligible overhead compared to the baseline.

Implement the Analytical Hierarchy Process (AHP) and K-Nearest Neighbor (KNN) Algorithms for Sales Classification

The Analytical Hierarchy Process (AHP) and K-Nearest Neighbor (KNN) algorithms are two algorithms that have proven efficient in various classification and prediction applications. This research examines the application of these two algorithms in the context of selling goods in PIM supermarkets. In this research, AHP and KNN are used to classify goods sold based on various criteria such as price, number of stock items sold, total sales amount. The research results show that KNN outperforms AHP in predicting the best-selling, best-selling and least-selling items based on sales in 2022 at PIM supermarkets. Based on this research, it can be concluded that the KNN algorithm is suitable for predicting the classification of goods sales in PIM Supermarkets. This research classifies sales of goods using the Analytical Hierarchy Process (AHP) and K-Nearest Neighbor (KNN) methods. This research uses 3 criteria. By using the value K=1, the experimental results show that the highest KNN has an accuracy of 38%, while AHP has an accuracy of 32%. Differences in accuracy results can be influenced by parameter settings and characteristics of the dataset used. Therefore, further analysis of these factors is needed to understand the performance differences between the two methods.

Blood Pressure and Heart Rate Measurement for Hypertension Classification Using the K-Nearest Neighbors Method Based on IoT

This research aims to develop a hypertension classification system based on measuring blood pressure and heart rate using the K-Nearest Neighbor (KNN) method and storing data on the Thingspeak cloud server. Hypertension is a major health problem that requires regular monitoring for effective prevention and management. This research created a tool that can measure blood pressure and heart rate using the KNN method as a classification. The data was collected using a sensor device integrated with Thingspeak for real-time data storage and analysis. The KNN method is used to classify measurement data into optimal, normal, prehypertension, grade 1 hypertension, and grade 2 hypertension categories. The implementation of data storage on the Thingspeak cloud server allows easy data access and efficient analysis, and supports continuous health monitoring. In conclusion, the system developed can be an effective tool in monitoring and classifying hypertension, and has the potential to be applied on a wider scale for public health management.

Analysis of Public Interest in Smartfren SIM Cards using the K-Nearest Neighbors Method

The use of Smartfren SIM cards is increasing along with the public's need for fast and stable internet services. However, a deep understanding of public interest in the SIM card is necessary to optimize marketing strategies and increase sales. Proper analysis can help companies identify potential target markets and develop effective marketing strategies. We chose the K-Nearest Neighbors method to analyze public interest in using Smartfren SIM cards. This study aims to develop and evaluate the K-Nearest Neighbors model in predicting public interest in using Smartfren SIM cards. This study uses a dataset containing information about Smartfren SIM card users. We divide the data into two sets: a training set for model building and a test set for evaluating model performance. We apply the K-Nearest Neighbors method to classify the data into two categories: interested and not interested. We evaluate the model performance using accuracy, precision, recall, and F1-score metrics. We present the evaluation results as a confusion matrix. The developed K-Nearest Neighbors model showed excellent performance with an accuracy of 94.29%, a precision of 94.20%, a recall of 100%, and an F1-score of 97.01%. These results indicate that the K-Nearest Neighbors model is effective in predicting people's interest in Smartfren SIM cards. The high recall value indicates that the model is able to identify all interested individuals without missing any, while the high precision value indicates that the model rarely makes false positive prediction errors. This study concludes that the K-Nearest Neighbors method is very effective for use in analyzing people's interest in using Smartfren SIM cards. We can rely on the developed model's strong performance for real-world applications in marketing strategies.

Implementing K-Nearest Neighbors (k-NN) Algorithm and Backward Elimination on Cardiotocography Datasets

Having a healthy baby is a dream for mothers. Unfortunately, high maternal and fetal mortality has become a vital problem that requires early risk detection for pregnant women. A cardiotocograph examination is necessary to maintain maternal and fetal health. One method that can solve this problem is classification. This research analyzes the optimal use of k values and distance measurements in the k-NN method. This research expects to become the primary reference for other studies examining the same dataset or developing k-NN. A selection feature is needed to optimize the classification method, particularly for improving accuracy results. This study used the cardiotocography dataset from cardiotocograph examinations related to fetal conditions. The cardiotocography dataset consisted of 2,126 records with 22 features and variables. It also had three classification classes, normal, suspect, and pathological, from the Universal Child Immunization Machine Learning Repository website. It employed the K-Nearest Neighbor (k-NN) method and the backward elimination feature with ordinary least squares regression. The test in this research applied the scenarios of three distance calculations, i.e., Euclidean distance, Manhattan distance, and Minkowski distance, as well as four variations of k values. Evaluation of each scenario indicated the accuracy of the confusion matrix and execution time. This study compared K-Nearest Neighbor (k-NN) and Backward Elimination methods with K-nearest neighbor (k-NN) without selection features. The best accuracy of the Backward Elimination and K-Nearest Neighbor (K-NN) methods was 91%, as was the K-Nearest Neighbor (k-NN) method without selection features. Both had similar k values (k = 3) and Manhattan distance. The backward elimination method reduced the number of features from 22 to 14. Meanwhile, the execution times of the Backward Elimination and K-Nearest Neighbor (k-NN) methods got better results as each distance averaged 26.54, 19.23, and 68.09 seconds. K-Nearest Neighbor (k-NN) execution times without selection features were 26.83, 19.39, and 68.84, respectively. In conclusion, backward elimination did not increase accuracy because it yielded the same accuracy. However, backward elimination and K-nearest Neighbor (k-NN) produced faster results, with differences of 29%, 16%, and 75%, respectively.

Analysis of Public Satisfaction Levels towards Hospital Services using the K-Nearest Neighbors Method (Case Study: XYZ Regional Public Hospital)

Analysis of Public Satisfaction Levels towards Hospital Services using the K-Nearest Neighbors Method (Case Study: XYZ Regional Public Hospital)

Analysis Of Public Interest In Smartfren SIM Cards Using The K-Nearest Neighbors Method

The use of Smartfren SIM cards is increasing along with the public's need for fast and stable internet services. However, a deep understanding of public interest in the SIM card is necessary to optimize marketing strategies and increase sales. Proper analysis can help companies identify potential target markets and develop effective marketing strategies. We chose the K-Nearest Neighbors method to analyze public interest in using Smartfren SIM cards. This study aims to develop and evaluate the K-Nearest Neighbors model in predicting public interest in using Smartfren SIM cards. This study uses a dataset containing information about Smartfren SIM card users. We divide the data into two sets: a training set for model building and a test set for evaluating model performance. We apply the K-Nearest Neighbors method to classify the data into two categories: interested and not interested. We evaluate the model performance using accuracy, precision, recall, and F1-score metrics. We present the evaluation results as a confusion matrix. The developed K-Nearest Neighbors model showed excellent performance with an accuracy of 94.29%, a precision of 94.20%, a recall of 100%, and an F1-score of 97.01%. These results indicate that the K-Nearest Neighbors model is effective in predicting people's interest in Smartfren SIM cards. The high recall value indicates that the model is able to identify all interested individuals without missing any, while the high precision value indicates that the model rarely makes false positive prediction errors. This study concludes that the K-Nearest Neighbors method is very effective for use in analyzing people's interest in using Smartfren SIM cards. We can rely on the developed model's strong performance for real-world applications in marketing strategies.

Comparison of Social Media BOT Functions Using the K-Nearest Neighbor Method Against User Satisfaction

Social media is currently an alternative medium in conveying messages in the form of news and can be used as a tool to exchange news from different places. Many people use social media to express opinions, express feelings, as well as experiences and things that can be of concern. In this study, the data processing used was the K-Nearest Neighbor Algorithm with the classification method as a media for comparing the functions of the two bots, namely the WhatsApp and Telegram bots. Using SPSS (Statistical Product and Serive Solutions) and Rapidminer as a place to perform calculations and analysis. Based on the results of testing data mining with Rapid Miner, the calculation results are obtained which will be used as information to support user satisfaction in using Social Media Bots. User satisfaction is found in WhatsApp Bot users 72.22% and Telegram Bot users 28.57%. Calculations are carried out with a data mining process obtained from the K-Nearest Neighbor algorithm to make it easier to find Bot user satisfaction on both WhatsApp and Telegram social media. The WhatsApp bot is the best choice and has several useful functions as a digital communication media tool on the Internet of Things.