Search

Virtual machine (VM) consolidation and migration that only consider current workload can result in excessive unnecessary migrations. To address this issue, a VM consolidation algorithm based on resource utilization prediction is proposed. An improved K-nearest neighbor (KNN) classification algorithm weighted by attribute inconsistency is proposed to predict the workload of both the host and the VMs. Firstly, two distributions are partitioned according to the neighboring relationship for comparing consistency. Then, an inconsistency evaluation function based on earth mover's distance (EMD) is designed to measure the inconsistency between the neighboring sample set of each sample under each attribute and the equivalent partition refined by the decision attribute. Finally, the inconsistency level of the neighboring samples is transformed into the importance of the corresponding attribute to implement the attribute weighting KNN classifier. When selecting the source host and target host for VM migration, both current and predicted overloads are considered to avoid unnecessary VM migrations. Simulation tests were performed with random and realistic workloads, and the results show that the proposed method can reduce the overall energy consumption of the host, while also reducing service level agreement (SLA) violations and VM migration. [ABSTRACT FROM AUTHOR]

• For the first time 100% accuracy for epileptic seizure detection using electroencephalogram signals is demonstrated. • Both balanced and unbalanced datasets are studied. • Wavelet extraction from the original EEG signal enhances the feature extraction and selection steps. • Automatic segmentation with algorithmic intelligence is used. • As opposed to Genetic algorithms fast GA is used to achieve fast analysis. The precise electroencephalogram (EEG) signal classification with the highest possible accuracy is a key goal in the brain-computer interface (BCI). Considering the complexity and nonstationary nature of the EEG signals, there is an urgent need for effective feature extraction and data mining techniques. Here, we introduce a novel pipeline based on Bat and genetic algorithms for feature construction and dimension reduction of EEG signals. After wavelet extraction and segmentation, the Bat algorithm identifies the most relevant features. We use these features and a genetic algorithm combined with a neural network method to automatically classify the segments of the epilepsy EEG signals. We also use available classification methods based on k-Nearest Neighbors or naïve Bayes for comparison purposes. The code distinguishes individual signals within various combinations of data obtained from healthy volunteers with open or closed eyes and patients suffering from epilepsy disorders during seizure-free periods or seizure activities. Compared to the previously introduced methods, our proposed framework demonstrates a superior balance of high accuracy and short runtime. The minimum achieved accuracies for balanced and unbalanced classes are 100% and 75.9%, respectively. This approach has the potential for direct applications in clinics, enabling accurate and rapid analysis of the epilepsy EEG signals obtained from patients. [ABSTRACT FROM AUTHOR]

Density peaks clustering (DPC) algorithm has been widely applied in many fields due to its innovation and efficiency. However, the original DPC algorithm and many of its variants choose Euclidean distance as local density and relative distance estimations, which affects the clustering performance on some specific shaped datasets, such as manifold datasets. To address the above-mentioned issue, we propose a density peak clustering algorithm with connected local density and punished relative distance (DPC–CLD–PRD). Specifically, the proposed approach computes the distance matrix between data pairs using the flexible connectivity distance metric. Then, it calculates the connected local density of each data point via combining the flexible connectivity distance measure and k-nearest neighbor method. Finally, the punished relative distance of each data point is obtained by introducing a connectivity estimation strategy into the distance optimization process. Experiments on synthetic, real-world, and image datasets have shown the effectiveness of the algorithm in this paper. [ABSTRACT FROM AUTHOR]

The suggested solution in this work makes use of the parallel processing capability of FPGA to enhance the efficiency of the K-Nearest Neighbor (KNN) algorithm on encrypted data. The suggested technique was assessed utilizing the breast cancer datasets and the findings indicate that the FPGA-based acceleration method provides significant performance improvements over software implementation. The Cheon–Kim–Kim–Song (CKKS) homomorphic encryption scheme is used for the computation of ciphertext. After extensive simulation in Python and implementation in FPGA, it was found that the proposed architecture brings down the computational time of KNN on ciphertext to a realistic value in the order of the KNN classification algorithm over plaintext. For the FPGA implementation, we used the Intel Agilex7 FPGA (AGFB014R24B2E2V) development board and validated the speed of computation, latency, throughput, and logic utilization. It was observed that the KNN on encrypted data has a computational time of 41.72 ms which is 80 times slower than the KNN on plaintext whose computational time is of 0.518 ms. The main computation time for CKKS FHE schemes is 41.72 ms. With our architecture, we were able to reduce the calculation time of the CKKS-based KNN to 0.85 ms by using 32 parallel encryption hardware and reaching 300 MHz speed. [ABSTRACT FROM AUTHOR]

This study presents a method to address the issue of burdensome computations in water resources optimization based on a hybrid algorithm derived from the firefly algorithm (FA) and the K-nearest neighbor (KNN) algorithm, herein named the FA-KNN algorithm. The FA-KNN algorithm introduced in this work is tested with three standard test problems (the Ackley, Rosenbrock, and Sphere problems), and with a reservoir operation problem that minimizes the relative agricultural water-supply deficit under baseline and climate-change conditions. The efficiency indexes of the reservoir system are calculated to evaluate the performance of the FA-KNN algorithm and its accuracy. The results demonstrate the operational policy obtained with the FA-KNN algorithm has better performance in terms of computational burden than the FA's. This work's findings establish the FA-KNN hybrid algorithm reduces the computational time by 60% with acceptable accuracy compared with the FA algorithm. The findings indicate a reduction in run-time of 99.5, 94, and 92% for solving the Ackley, Rosenbrock, and Sphere test problems achieved with the FA-KNN algorithm while maintaining a high level of accuracy when contrasted with solutions derived from both deterministic methodologies and the FA approach. The volumetric reliability and flexibility in the reservoir problem calculated under the baseline conditions outperformed those obtained with the climate change conditions by 10 and 3.5%, respectively. Moreover, a notable discrepancy emerged in terms of the main simulator's invocation frequency between the FA-KNN and FA methods (the former exhibited a mere 0.3 ratio compared to the latter). The application of the FA-KNN approach yielded a reduction exceeding 60% in run-time for the reservoir problem. [ABSTRACT FROM AUTHOR]

The main purpose of this study is to forecast the countries' socio-economic performance with the fewest possible parameters. To do this, we propose a model consisting of methods from Multi-Criteria Decision Making and Machine Learning. Since the existence of different classifications of countries and several socioeconomic parameters, it becomes difficult to make a prediction of their belonging group and compare countries based on these parameters. Using the Grey Relational Analysis and the Critic method, we classify the countries into four different subgroups based on several socio-economic dimensions. K-Nearest Neighbor (K-NN) algorithm with basic macro-economic parameters is implemented to predict the countries' socioeconomic groups. The results rank the countries according to their socio-economic performance and predict the countries' development levels for the future. The main findings indicated that the proposed approach can be used for similar research questions. The highest prediction percentages are accurate for small values of k. This study provides a convenient and effective method for grouping countries at different levels of development using basic economic parameters and provides a simple and practical method to predict the belonging group. [ABSTRACT FROM AUTHOR]

To address the problem of low quality of the outlier detection results caused by the irregular spatial distribution of crowdsourced bathymetric data, an intelligently optimized 3D-DBSCAN method is hereby proposed for detecting outliers in crowdsourced bathymetric data. Firstly, the potential of the classic DBSCAN algorithm in clustering irregularly distributed data has been explored for three-dimensional space and upgraded to the 3D-DBSCAN algorithm. Meanwhile, three key parameters affecting the 3D-DBSCAN algorithm results have been determined, namely, the minimum number of objects in the neighborhood, the horizontal neighborhood radius and the vertical neighborhood radius. Then, the abilities of the K-nearest neighbor method and the genetic algorithm in setting the initial range of parameters and searching for optimal values have been comprehensively utilized. The optimal combination solution vector of the key parameters of 3D-DBSCAN under different distribution conditions has been adaptively acquired by defining and calculating the silhouette coefficient index. Finally, spatial clustering analysis on crowdsourced bathymetric data and intelligently detect outliers has been conducted by using the optimal combination solution vector as the input of the 3D-DBSCAN model parameters. The experimental results show that the proposed method can not only detect outliers in crowdsourced bathymetric data after adaptively optimizing the parameters, but also detect outliers by analyzing the distance relationship between points in the 3D space, thereby overcoming the limitations of traditional methods in detecting water depth anomalies at specific positions and irregularly distributed anomalies. [ABSTRACT FROM AUTHOR]

In the majority of data mining tasks, feature selection serves as an essential pre-processing step. The most important attributes are selected to lower the dimensionality reduction of data set and enhance the precision of classification. Natural heuristic algorithms are extensively employed in the realm of encapsulated feature selection. Based on the wrapper feature selection method, seven natural heuristic algorithms are used to solve feature selection problems and perform performance comparison, which include Slime Mold Algorithm (SMA), Whale Optimization Algorithm (WOA), Harris Hawks Optimization Algorithm (HHO), Marine Predator Algorithm (MPA), Butterfly Optimization Algorithm (BOA), Cuckoo Search (CS) and Firefly Algorithm (FA). At the same time, performance tests are carried out on 21 standard UCI data sets to verily the functionality of various algorithms, and the convergence curves and accuracy boxplots of 7 natural heuristic algorithms on 21 data sets are given. The simulation outcomes were assessed utilizing the mean and standard deviation of fitness, as well as the number of chosen features, and the running time, with the optimal value in bold. By comparing the comprehensive performance indexes, MPA obtained the maximum mean fitness value in most data sets (16 data sets), followed by FA (6 data sets). SMA obtained the best performance and finds the minimum eigenvalues (20 data sets) in multiple data sets and has an advantage in computing time. [ABSTRACT FROM AUTHOR]

Many studies have shown that long-chain noncoding RNAs (lncRNAs) are involved in a variety of biological processes such as post-transcriptional gene regulation, splicing, and translation by combining with corresponding proteins. Predicting lncRNA-protein interactions is an effective approach to infer the functions of lncRNAs. The paper proposes a new computational model named LPI-IBWA. At first, LPI-IBWA uses similarity kernel fusion (SKF) to integrate various types of biological information to construct lncRNA and protein similarity networks. Then, a bounded matrix completion model and a weighted k -nearest known neighbors algorithm are utilized to update the initial sparse lncRNA-protein interaction matrix. Based on the updated lncRNA-protein interaction matrix, the lncRNA similarity network and the protein similarity network are integrated into a heterogeneous network. Finally, an improved Bi-Random walk algorithm is used to predict novel latent lncRNA-protein interactions. 5-fold cross-validation experiments on a benchmark dataset showed that the AUC and AUPR of LPI-IBWA reach 0.920 and 0.736, respectively, which are higher than those of other state-of-the-art methods. Furthermore, the experimental results of case studies on a novel dataset also illustrated that LPI-IBWA could efficiently predict potential lncRNA-protein interactions. • LPI-IBWA is an effective computational model to predict lncRNA-protein interactions. • Similarity kernel fusion is used to integrate heterogenous biological similarities. • Bounded matrix completion updates lncRNA-protein interaction matrix. • Bi-Random walk on lncRNA-protein interactions network predicts novel interactions. [ABSTRACT FROM AUTHOR]

The quality of the sample plays a vital role in developing accurate models using machine learning. This aspect is equally important when evaluating regional landslide susceptibility using machine learning. Previous studies have mostly employed random generation methods to select samples, which often fail to select representative samples. Therefore, this study proposes the KK-sampling method, which uses K-means and KNN algorithms to analyze relevant attributes of the study area and select samples. To evaluate the effectiveness of the proposed method, this study employed MLP, RF, and XGBoost models in conjunction with the KK-sampling method, with Zhong County, Chongqing serving as a case study. The results indicate that the KK-sampling method significantly improves the stability and accuracy of the model. Additionally, this study analyzed the importance of landslide factors in Zhong County using SHAP values. The findings provide a reference for establishing a reasonable and effective landslide susceptibility model in the region. [ABSTRACT FROM AUTHOR]

Algorithms for answering the k nearest-neighbor (k-NN) query are widely used for queries in spatial databases and for distance classification of a group of query points against a reference dataset to derive the dominating feature class. GPU devices have significantly more processing cores than CPUs and faster device memory than the main memory accessed by CPUs, thus, providing higher computing power for processing demanding queries like the k-NN. However, since device and/or main memory may not be able to host an entire, rather big, reference and query datasets, storing these datasets in a fast secondary device, like a solid state disk (SSD), and partially retrieve the required, at each stage, partitions is, in many practical cases, a feasible solution. We propose and implement the first GPU-based algorithms for processing the k-NN query for big reference and query spatial data stored on SSDs. Based on 3d synthetic and real big spatial data, we experimentally compare these algorithms and highlight the most efficient algorithmic variation. This variation utilizes a CUDA feature known as Concurrent Kernel Execution, to further improve its performance. [ABSTRACT FROM AUTHOR]

Due to atmospheric scattering, solar radiation, and other factors, the Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) product data suffer from a substantial amount of background noise. This poses a significant challenge when attempting to directly utilize the raw data. Consequently, data denoising becomes an indispensable preprocessing step for its subsequent applications, such as the extraction of forest structure parameters and ground elevation data. While the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is currently the most widely used method, it remains susceptible to complexities arising from terrain, low signal-to-noise ratio (SNR), and input parameter variations. This paper proposes an efficient Multi-Level Auto-Adaptive Noise Filter (MLANF) algorithm based on photon spatial density. Its purpose is to extract signal photons from ICESat-2 terrestrial data of different ground cover types. The algorithm follows a two-step process. Firstly, random noise photons are removed from the upper and lower regions of the signal photons through a coarse denoising process. Secondly, in the fine denoising step, the K-Nearest Neighbor (KNN) algorithm selects the K photons to calculate the slope along the track. The calculated slope is then used to rotate the direction of the searching neighborhood in the DBSCAN algorithm. The proposed algorithm was tested in eight datasets of four surface types: forest, grassland, desert, and urban, and the extraction results were compared with those from the ATL08 datasets and the DBSCAN algorithm. Based on the ground-truth signal photons obtained by visual inspection, the classification precision, recall, and F-score of our algorithm, as well as two other algorithms, were calculated. The MLANF could achieve a good balance between classification precision (97.48% averaged) and recall (97.96% averaged). Its F-score (97.69% averaged) was higher than that of the other two methods. This demonstrates that the MLANF algorithm successfully obtained a continuous surface profile from ICESat-2 datasets with different surface cover types, significant topographic relief, and low SNR. [ABSTRACT FROM AUTHOR]

Given that fingerprint localization methods can be effectively modeled as supervised learning problems, machine learning has been employed for indoor localization tasks based on fingerprint methods. However, it is often challenging for popular machine learning models to effectively capture the unstructured data features inherent in fingerprint data that are generated in diverse propagation environments. In this paper, we propose an indoor localization algorithm based on a high-order graph neural network (HoGNNLoc) to enhance the accuracy of indoor localization and improve localization stability in dynamic environments. The algorithm first designs an adjacency matrix based on the spatial relative locations of access points (APs) to obtain a graph structure; on this basis, a high-order graph neural network is constructed to extract and aggregate the features; finally, the designed fully connected network is used to achieve the regression prediction of the location of the target to be located. The experimental results on our self-built dataset show that the proposed algorithm achieves localization accuracy within 1.29 m at 80% of the cumulative distribution function (CDF) points. The improvements are 59.2%, 51.3%, 36.1%, and 22.7% compared to the K-nearest neighbors (KNN), deep neural network (DNN), simple graph convolutional network (SGC), and graph attention network (GAT). Moreover, even with a 30% reduction in fingerprint data, the proposed algorithm exhibits stable localization performance. On a public dataset, our proposed localization algorithm can also show better performance. [ABSTRACT FROM AUTHOR]

The authors propose to use the K-nearest neighbor (K-NN) algorithm to process the probe signals from the previously proposed Brillouin optical time-domain reflectometry (BOTDR) fiber sensor called the differential cross-spectrum BOTDR (DCS-BOTDR), which features a high spatial resolution. Widening of the Brillouin gain spectrum (BGS) when the pulse duration TL is set shorter than the phonon lifetime (~10 ns) is a key problem associated with the DCS-BOTDR. Although the spectrum distortion is reduced also, deterioration in the Brillouin frequency shift (BFS) resolution is observed, which is caused by the broadening of the Brillouin spectrum. On the other hand, while the TL duration exceeding the steady state of the Brillouin signal narrows the spectrum, sidelobes are produced within the steady-state range, which result in the deterioration of the BFS resolution. This limitation is observed only in the DCS-BOTDR, although it is capable of measuring with a high spatial resolution. The experimental data obtained for a fiber with a length of ~400 m and with optimized TL duration in the temperature range of 40–80°C were used in our model to extract the Brillouin temperature coefficient CT. Ideal BGSs were then constructed in the training phase by simulation with different linewidths of 50–70 MHz in order to train the K-NN model with due account for the linewidth variation caused by the difference in the conditions between the training and testing phases. The model was thereby made flexible for various fiber conditions. Experimental data for the sensing of a ~3.6-km-long fiber with TL = 60 ns were used in the testing phase to obtain the temperature distribution. By employing K-NN, the accuracy in determining the temperature for TL = 60 ns has been improved to approximately 2.77°C. Therefore, the K-NN model can be an excellent alternative tool for processing BGSs measured by the DCS-BOTDR and obtaining the temperature distribution along the fiber. [ABSTRACT FROM AUTHOR]

Soil discrimination is one of the most critical forensic analyses aiming to establish a link between a suspect and a crime scene or victim. Preliminary studies have indicated that ultra-performance liquid chromatography (UPLC) is useful in acquiring organic profiles of soils. However, the resulting chromatograms are often imperfect but consist of fluctuated baseline and overlapped peaks. Hence, this work aims to compare the performances of two derivative algorithms, i.e., Gap-Segment (GS) and Savitzky–Golay (SG) algorithms, in improving the UPLC chromatograms of soils for forensic investigation purposes. A set of 45 chromatograms was prepared by analyzing 15 Malaysian soil samples in triplicate by using a UPLC-photodiode array detector system. The capability of the two algorithms in discriminating the chromatograms from five geographical origins was studied based on predictive capability of K-nearest neighbor algorithm. Results showed that the GS algorithm slightly outperformed the SG algorithm. In conclusion, lower polynomial and differentiation order are preferred for improving the quality of the UPLC chromatograms of soils. The study provides an understanding of the relative merits between Gap-Segment and Savitzky–Golay algorithms in preprocessing UPLC data. [ABSTRACT FROM AUTHOR]

At present, the gap of agricultural talents in China is continuously widening, and most enterprises lack agricultural core talents, which has caused great impact on the social economy. To solve this problem, an improved AHP-KNN algorithm is proposed by combining the analytic hierarchy process (AHP) and the optimized K-nearest neighbor algorithm, and an agricultural talent training model is proposed based on this algorithm. The results show that the classification accuracy and classification time of the improved AHP-KNN algorithm are 96.2% and 27.5 seconds, respectively, both of which are superior to the comparison algorithm. The result shows that the classification accuracy of agricultural talents can be improved by using this algorithm. Therefore, the model can be used to classify agricultural talents with the same characteristics into one class, carry out targeted training, and train all-round agricultural talents efficiently and quickly, so as to improve the serious shortage of agricultural talents at present. [ABSTRACT FROM AUTHOR]

Graph-based clustering performs efficiently for identifying clusters in local and nonlinear data Patterns. The existing methods face the problem of parameter selection, such as the setting of k of the k-nearest neighbor graph and the threshold in noise detection. In this paper, a non-parametric clustering algorithm (NonPC) is proposed to tackle those inherent limitations and improve clustering performance. The weighted natural neighbor graph (wNaNG) is developed to represent the given data without any prior knowledge. What is more, the proposed NonPC method adaptively detects noise data in an unsupervised way based on some attributes extracted from wNaNG. The algorithm works without preliminary parameter settings while automatically identifying clusters with unbalanced densities, arbitrary shapes, and noises. To assess the advantages of the NonPC algorithm, extensive experiments have been conducted compared with some classic and recent clustering methods. The results demonstrate that the proposed NonPC algorithm significantly outperforms the state-of-the-art and well-known algorithms in Adjusted Rand index, Normalized Mutual Information, and Fowlkes-Mallows index aspects. [ABSTRACT FROM AUTHOR]

The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) is equipped with the Advanced Topographic Laser Altimeter System (ATLAS), which can achieve high-precision ground detection. However, due to its low pulse energy and high sensitivity, it is also affected by noise when obtaining data. Therefore, data denoising is critical to the subsequent processing and application. In this study, a multilevel filtering algorithm is proposed to denoise the daytime photon cloud data with high background noise. Firstly, the Random Sample Consensus (RANSAC) algorithm is used to roughly denoise the daytime photon cloud data with high background noise, and a signal point cloud buffer is established to remove most of the noise points. Subsequently, the horizontal continuity parameter is calculated based on the photon cloud data following the rough denoising process. Based on this parameter, the shape and size of the search domain of the results of the subsequent fine denoising algorithm are adaptively improved. Finally, three filtering directions (a horizontal direction, an intra-group unified direction, and an adaptive direction for each photon) are proposed, and a hybrid algorithm combining the Ordering Points to Identify the Clustering Structure (OPTICS) density clustering algorithm and the Relative Neighboring Relationship K-nearest neighbors-based noise removal (RNR−KNNB) algorithm is employed to accurately denoise the photon cloud data in the three filtering directions. Furthermore, the RANSAC algorithm based on a sliding overlap window is used to remove outliers for the weak beam fine denoising photon cloud data. The results indicate that, for the strong beams, the denoising accuracy of the multilevel filtering algorithm in the three filtering directions is comparable (Rs ≥ 0.96, F ≥ 0.67), and they are all better than that of the ATL08 algorithm (Rs/Rn/p/F = 0.85/0.67/0.52/0.65). For weak beams, the denoising accuracy of the multilevel filtering algorithm in the horizontal direction and the intra-group unified direction is similar (Rs = 0.92, F = 0.69), and it is superior to the denoising results in the adaptive direction of each photon and the ATL08 algorithm (Rs/Rn/p/F = 0.94/0.84/0.51/0.65, 0.88/0.87/0.55/0.67, respectively). For strong and weak beams, the p-value and F-value of the denoising results of multilevel filtering algorithms in three different filtering directions increase with the increase of SNR value. It is demonstrated that SNR is an important factor affecting the denoising results of algorithms. The multilevel filtering algorithm proposed in the study can effectively achieve precise denoising of daytime photon cloud data with high background noise, and the three different filtering directions have different effects on the denoising results of strong and weak beam photon cloud data. This can provide technical and methodological guidance for subsequent photon cloud data filtering processing. [ABSTRACT FROM AUTHOR]

Predicting attacks in Android malware devices using machine learning for recommender systems-based IoT can be a challenging task. However, it is possible to use various machine-learning techniques to achieve this goal. An internet-based framework is used to predict and recommend Android malware on IoT devices. As the prevalence of Android devices grows, the malware creates new viruses on a regular basis, posing a threat to the central system's security and the privacy of the users. The suggested system uses static analysis to predict the malware in Android apps used by consumer devices. The training of the presented system is used to predict and recommend malicious devices to block them from transmitting the data to the cloud server. By taking into account various machine-learning methods, feature selection is performed and the K-Nearest Neighbor (KNN) machine-learning model is proposed. Testing was carried out on more than 10,000 Android applications to check malicious nodes and recommend that the cloud server block them. The developed model contemplated all four machine-learning algorithms in parallel, i.e., naive Bayes, decision tree, support vector machine, and the K-Nearest Neighbor approach and static analysis as a feature subset selection algorithm, and it achieved the highest prediction rate of 93% to predict the malware in real-world applications of consumer devices to minimize the utilization of energy. The experimental results show that KNN achieves 93%, 95%, 90%, and 92% accuracy, precision, recall and f1 measures, respectively. [ABSTRACT FROM AUTHOR]

Heavy losses in the wind power generation system are incurred when the wind turbine blades are severely damaged beyond repair. Therefore, an immediate repair at an initial stage is an economical method. There are limitations and difficulties in diagnosing initial damage to wind turbine blades. For example, the measurement cannot be performed during actual operation, the diagnostic machine learning model calculations are too complex, and so on. This study measured the noise signals in the operation of wind turbines. The k-nearest neighbor (k-NN) of supervised learning was used as the diagnostic method. The calculation of generalized fractal dimensions (GFDs) was used as diagnostic feature selection. The result shows an accuracy of 98.9%. Compared to other algorithms, the k-NN algorithm is simple and easy to understand and implement. To reduce the amount of calculation of the machine learning model, the optimum numerical combination of three major parameters is found in this study. These major parameters are (1) scale index of GFDs, (2) number of neighbor points in the algorithm, and (3) range formula. According to the findings, the k-NN algorithm model can achieve high accuracy. The optimum numerical combination of three major parameters provides a rapid, convenient, and efficient diagnostic method. [Display omitted] [ABSTRACT FROM AUTHOR]

Along with the IoT technology, the importance of indoor positioning is increasing, but the accuracy of the traditional fingerprint positioning algorithm is negatively affected by the complex indoor environment. This issue of low indoor spatial geolocation localization accuracy when the signal is collected away from the present stage occurs due to the signal instability of the iBeacon in the traditional fingerprint localization algorithm, which generates a variety of factors such as object blocking and reflection, multipath effect, etc., as well as the scarcity of reference fingerprint data points. In response, this study proposes an inverse distance-weighted optimization WKNN algorithm for indoor localization based on the GM(1,1) model. By implementing GM(1,1) model pre-process leveling, the original fingerprint library was reconstructed into a large-capacity fingerprint database using the inverse distance-weighted interpolation method. The local inverse distance-weighted interpolation was used for interpolation, combined with the WKNN algorithm to complete the coordinate solution in real time. This effectively solved the issue of low localization accuracy caused by the large fluctuation of the received signal strength (RSS) sampling measurement data and the existence of few reference fingerprint datapoints in the fingerprint database. The results show that this algorithm reduced the average positioning error by 5.9% compared with ordinary kriging (OK) interpolation leveling and reduced the average positioning error by 18.2% compared with the indoor spatial location accuracy of the original fingerprint database, which can effectively improve the positioning accuracy and provide technical support for indoor location and navigation services. [ABSTRACT FROM AUTHOR]

Quality-related prediction in the continuous-casting process is important for the quality and process control of casting slabs. As intelligent manufacturing technologies continue to evolve, numerous data-driven techniques have been available for industrial applications. This case study was aimed at developing a machine-learning algorithm, capable of predicting slag inclusion defects in continuous-casting slabs, based on process condition sensor data. A large dataset consisting of sensor data from nearly 7300 casting samples has been analyzed, with the empirical mode decomposition (EMD) algorithm utilized to process the multi-modal time series. The following machine-learning algorithms have been examined: K-Nearest neighbors, support vector classifier (linear and nonlinear kernels), decision trees, random forests, AdaBoost, and Artificial Neural Networks. Four over-sampling or under-sampling algorithms have been adopted to solve imbalanced data distribution. In the experiment, the optimized random forest outperformed other machine-learning algorithms in terms of recall and ROC AUC, which could provide valuable insights for quality control. [ABSTRACT FROM AUTHOR]

The molecularly imprinted polymers are artificial polymers that, during the synthesis, create specific sites for a definite purpose. These polymers due to their characteristics such as stability, easy of synthesis, reproducibility, reusability, high accuracy, and selectivity have many applications. However, the variety of the functional monomers, templates, solvents, and synthesis conditions like pH, temperature, the rate of stirring, and time, limit the selectivity of imprinting. The Practical optimization of the synthetic conditions has many drawbacks, including chemical compound usage, equipment requirements, and time costs. The use of machine learning (ML) for the prediction of the imprinting factor (IF), which indicates the quality of imprinting is a very interesting idea to overcome these problems. The ML has many advantages, for example a lack of human error, high accuracy, high repeatability, and prediction of a large amount of data in the minimum time. In this research, ML was used to predict the IF using non-linear regression algorithms, including classification and regression tree, support vector regression, and k-nearest neighbors, and ensemble algorithms, like gradient boosting (GB), random forest, and extra trees. The data sets were obtained practically in the laboratory, and inputs, included pH, the type of the template, the type of the monomer, solvent, the distribution coefficient of the MIP (KMIP), and the distribution coefficient of the non-imprinted polymer (KNIP). The mutual information feature selection method was used to select the important features affecting the IF. The results showed that the GB algorithm had the best performance in predicting the IF, and using this algorithm, the maximum R2 value (R2 = 0.871), and the minimum mean absolute error (MAE = − 0.982), and mean square error were obtained (MSE = − 2.303). [ABSTRACT FROM AUTHOR]

Electricity theft has caused massive losses and damage to electricity utilities. The damage affects the electricity supply's quality and increases the generation load. The losses happen not only for the electricity utilities but also affect the legitimate users who have to pay excessive electricity bills. That is why the method to detect electricity theft is indispensable. Recently, machine learning algorithms have been used to develop a model for detecting electricity theft. However, most algorithms have problems due to imbalanced data, overfitting issues, and lack of data. Therefore, this paper proposes a solution that implements the oversampling technique to address the problems and increase the developed model's accuracy. It is used to perform oversampling on the imbalanced dataset. Our proposed method consists of a pre-processing step to remove empty values and extract several parameters. After that, the oversampling technique is performed on the result of the pre-processing step. The logistic regression model combined with the oversampling techniques shows the best performance results on the developed model of electricity theft detection based on the state electricity company customers. The experiment shows that the proposed method, logistic regression combined with the synthetic minority oversampling technique, shows superior performance in terms of the accuracy of the training data and data testing, precision, recall, and F1-scores of 98.97%, 98.7%, 95%, 99%, and 97%, respectively. Moreover, the experiment also shows that the proposed solution outperforms existing methods. [ABSTRACT FROM AUTHOR]

Collecting and analyzing data from all devices to improve the efficiency of business processes is an important task of Industrial Internet of Things (IIoT). In the age of data explosion, extensive text data generated by the IIoT have given birth to a variety of text representation methods. The task of text representation is to convert the natural language to a form that computer can understand with retaining the original semantics. However, these methods are difficult to effectively extract the semantic features among words and distinguish polysemy in natural language. Combining the advantages of convolutional neural network (CNN) and variational autoencoder (VAE), this paper proposes an intelligent CNN-VAE text representation algorithm as an advanced learning method for social big data within next-generation IIoT, which help users identify the information collected by sensors and perform further processing. This method employs the convolution layer to capture the local features of the context and uses the variational technique to reconstruct feature space to make it conform to the normal distribution. In addition, the improved word2vec model based on topical word embedding (TWE) is utilized to add topical information to word vectors to distinguish polysemy. This paper takes the social big data as an example to illustrate the way of the proposed algorithm applied in the next-generation IIoT and utilizes Cnews dataset to verify the performance of proposed method with four evaluating metrics (i.e., recall, accuracy, precision, and F1-score). Experimental results indicate that the proposed method outperforms word2vec-avg and CNN-AE in K-nearest neighbor (KNN), random forest (RF), and support vector machine (SVM) classifiers and distinguishes polysemy effectively. [ABSTRACT FROM AUTHOR]

Geographically weighted regression (GWR) is a classical method for estimating nonstationary relationships. Notwithstanding the great potential of the model for processing geographic data, its large-scale application still faces the challenge of high computational costs. To solve this problem, we proposed a computationally efficient GWR method, called K-Nearest Neighbors Geographically weighted regression (KNN-GWR). First, it utilizes a k-dimensional tree (KD tree) strategy to improve the speed of finding observations around the regression points, and, to optimize the memory complexity, the submatrices of neighbors are extracted from the matrix of the sample dataset. Next, the optimal bandwidth is found by referring to the spatial clustering relationship explained by K-means. Finally, the performance and accuracy of the proposed KNN-GWR method was evaluated using a simulated dataset and a Chinese house price dataset. The results demonstrated that the KNN-GWR method achieved computational efficiency thousands of times faster than existing GWR algorithms, while ensuring accuracy and significantly improving memory optimization. To the best of our knowledge, this method was able to run hundreds of thousands or millions of data on a standard computer, which can inform improvement in the efficiency of local regression models. [ABSTRACT FROM AUTHOR]

In order to limit the spread of the novel betacoronavirus (SARS-CoV-2), it is necessary to detect positive cases as soon as possible and isolate them. For this purpose, machine-learning algorithms, as a field of artificial intelligence, have been recognized as a promising tool. The aim of this study was to assess the utility of the most common machine-learning algorithms in the rapid triage of children with suspected COVID-19 using easily accessible and inexpensive laboratory parameters. A cross-sectional study was conducted on 566 children treated for respiratory diseases: 280 children with PCR-confirmed SARS-CoV-2 infection and 286 children with respiratory symptoms who were SARS-CoV-2 PCR-negative (control group). Six machine-learning algorithms, based on the blood laboratory data, were tested: random forest, support vector machine, linear discriminant analysis, artificial neural network, k-nearest neighbors, and decision tree. The training set was validated through stratified cross-validation, while the performance of each algorithm was confirmed by an independent test set. Random forest and support vector machine models demonstrated the highest accuracy of 85% and 82.1%, respectively. The models demonstrated better sensitivity than specificity and better negative predictive value than positive predictive value. The F1 score was higher for the random forest than for the support vector machine model, 85.2% and 82.3%, respectively. This study might have significant clinical applications, helping healthcare providers identify children with COVID-19 in the early stage, prior to PCR and/or antigen testing. Additionally, machine-learning algorithms could improve overall testing efficiency with no extra costs for the healthcare facility. [ABSTRACT FROM AUTHOR]

Computations that mimic nature are known as nature-inspired computing. Nature presents a wealthy source of thoughts and ideas for computing. The use of natural galvanized techniques has been found to provide machine solutions to complex problems. One of the challenging issues among researchers is high-dimensional data which contains a large number of unwanted, redundant, and irrelevant features. These redundant or unwanted features reduce the accuracy of machine learning models. Therefore, to solve this problem nowadays metaheuristic techniques are being used. The paper presents both surveys as well as comparison of five metaheuristic algorithms for feature selection. A wrapper-based feature selection approach using five nature-inspired techniques for feature selection has been applied. The binary version of the five swarm-based nature-inspired algorithms (NIAs), namely particle swarm optimization, whale optimization algorithm (WOA), grey wolf optimization (GWO), firefly algorithm, and bat algorithm. WOA and GWO are recent algorithms used for finding optimal feature subsets when there is no empirical information. The S-shape transfer function has been used to convert the continuous value to binary form and K-nearest neighbor is used to calculate the classification accuracy of selected feature subsets. To validate the results of the selected NIAs eleven benchmark datasets from the UCI repository are used. The strength of each NIA has been verified using a nonparametric test called the Friedman rank and Holm test. p value obtained shows that WOA is statistically significant and performs better than other models. [ABSTRACT FROM AUTHOR]

Due to subjectivity in refereeing, the results of race walking are often questioned. To overcome this limitation, artificial-intelligence-based technologies have demonstrated their potential. The paper aims at presenting WARNING, an inertial-based wearable sensor integrated with a support vector machine algorithm to automatically identify race-walking faults. Two WARNING sensors were used to gather the 3D linear acceleration related to the shanks of ten expert race-walkers. Participants were asked to perform a race circuit following three race-walking conditions: legal, illegal with loss-of-contact and illegal with knee-bent. Thirteen machine learning algorithms, belonging to the decision tree, support vector machine and k-nearest neighbor categories, were evaluated. An inter-athlete training procedure was applied. Algorithm performance was evaluated in terms of overall accuracy, F1 score and G-index, as well as by computing the prediction speed. The quadratic support vector was confirmed to be the best-performing classifier, achieving an accuracy above 90% with a prediction speed of 29,000 observations/s when considering data from both shanks. A significant reduction of the performance was assessed when considering only one lower limb side. The outcomes allow us to affirm the potential of WARNING to be used as a referee assistant in race-walking competitions and during training sessions. [ABSTRACT FROM AUTHOR]

Background: Since the onset of the COVID-19 pandemic in early 2020, the importance of timely and effective assessment of mental well-being has increased dramatically. Machine learning (ML) algorithms and artificial intelligence (AI) techniques can be harnessed for early detection, prognostication and prediction of negative psychological well-being states. Methods: We used data from a large, multi-site cross-sectional survey consisting of 17 universities in Southeast Asia. This research work models mental well-being and reports on the performance of various machine learning algorithms, including generalized linear models, k-nearest neighbor, naïve Bayes, neural networks, random forest, recursive partitioning, bagging, and boosting. Results: Random Forest and adaptive boosting algorithms achieved the highest accuracy for identifying negative mental well-being traits. The top five most salient features associated with predicting poor mental well-being include the number of sports activities per week, body mass index, grade point average (GPA), sedentary hours, and age. Conclusions: Based on the reported results, several specific recommendations and suggested future work are discussed. These findings may be useful to provide cost-effective support and modernize mental well-being assessment and monitoring at the individual and university level. [ABSTRACT FROM AUTHOR]

Admission to an appropriate college is an important process. The method of enquiry is an important part of it. As a result, students gather data, analyse it, and attempt to predict which college is best for them to attend. In general, students find it useful to research the trends in fee structures, academic staff numbers, multi-national companies tie up, and placements received by students during that academic year. K-Nearest Neighbors (k-NN) is one of the simplest learning algorithms used in supervised learning approaches. It makes assumptions about the similarity between new cases or data and available cases and places new cases in the category that most closely resembles the available categories. In this paper, the authors have tried to identify or predict the category or class of a particular dataset using the k-NN algorithm. [ABSTRACT FROM AUTHOR]

Considering the low indoor positioning accuracy and poor positioning stability of traditional machine-learning algorithms, an indoor-fingerprint-positioning algorithm based on weighted k-nearest neighbors (WKNN) and extreme gradient boosting (XGBoost) was proposed in this study. Firstly, the outliers in the dataset of established fingerprints were removed by Gaussian filtering to enhance the data reliability. Secondly, the sample set was divided into a training set and a test set, followed by modeling using the XGBoost algorithm with the received signal strength data at each access point (AP) in the training set as the feature, and the coordinates as the label. Meanwhile, such parameters as the learning rate in the XGBoost algorithm were dynamically adjusted via the genetic algorithm (GA), and the optimal value was searched based on a fitness function. Then, the nearest neighbor set searched by the WKNN algorithm was introduced into the XGBoost model, and the final predicted coordinates were acquired after weighted fusion. As indicated in the experimental results, the average positioning error of the proposed algorithm is 1.22 m, which is 20.26–45.58% lower than that of traditional indoor positioning algorithms. In addition, the cumulative distribution function (CDF) curve can converge faster, reflecting better positioning performance. [ABSTRACT FROM AUTHOR]