Search

In the present study, plant-scale process integration was applied to an Indian paper mill using the water cascade analysis (WCA) technique. Three limiting constraints, chemical oxygen demand (COD), total dissolved solids (TDS), and adsorbable organic halides (AOX), were considered for the study. A nearest neighbor algorithm was used to distribute the freshwater and recycled water among the plant operations. It was found that the limiting critical constraint depends upon the types of processes and streams involved in the integration. The limiting critical constraint can differ for different sections of the same industry, and can differ in different schemes of integration. After process integration, a 55.6% reduction in effluent flow, a 36% reduction in COD, and a 73% reduction in AOX were observed. After process integration, a 35.21% reduction in pollution costs can be achieved and, assuming the average production of the mill to be 225 tons per day, a savings of Indian rupees (INR) 1.73 per kg of paper produced can be achieved by employing process integration. The water cess was calculated as INR 3024.77 per day without integration for the sections that were considered for integration, while after integration, a 41.53% savings in the form of water cess was calculated. [Copyright &y& Elsevier]

This paper combines the digital twin system modeling method to conduct an in-depth study and analysis of graph-theoretic combinatorial optimization. This paper provides new ideas and approaches for optimal numerical analysis work by studying the digital twin modeling method that integrates digital modeling and graph theory combination, provides theoretical support for safe, stable, and economic operation of the system, proposes a solution for digital twin model based on big data platform, focuses on the nearest neighbor propagation (AP) and graph theory combination, solves the digital twin real-time monitoring data asynchronous, incomplete problem, and applies the algorithm to the digital twin model based on the big data platform for data preprocessing to achieve better results. This paper also presents a web-based digital twin system based on intelligent practical needs, analysis, and comparison of existing models, combined with digital twin technology, detailing the differences and connections between the various levels of numerical analysis and the implementation of this data in various fields, such as user management, equipment health management, product quality management, and workshop 3D navigation and detailed modeling of the digital twin system based on this numerical analysis to realize remote online monitoring, analysis, and management. In this paper, for the numerical analysis process, firstly, the key technologies of modeling and simulation operation control of production line based on digital twin are studied, and the rapid response manufacturing system based on a digital twin is designed and validated. Secondly, a scheduling technology framework for capacity simulation evaluation and optimization is established, and batching optimization, outsourcing decision, and rolling scheduling techniques are thus proposed to form a batching optimization algorithm based on priority rules, which realizes batching processing, outsourcing decision, and rolling scheduling of production orders to optimize equipment utilization and capacity. Finally, digital twin-based modeling is designed, and the validation results demonstrate the system's superior performance in achieving information interaction between physical and virtual production lines, optimization of numerical analysis, and display of results. [ABSTRACT FROM AUTHOR]

When recognizing targets by unmanned aerial vehicles (UAVs), problems such as small size, dense dispersion, and complex background are likely to occur, resulting in low recognition rates. In order to solve the above problems, this work proposed a lightweight small target detection algorithm based on the YOLOv8n: STD-YOLOv8 algorithm. First, the regression problem of small targets in different training periods was optimized, the penalty term in the original loss was improved, and a new LIoU loss function was proposed, so that the size of the penalty term could be dynamically adjusted before and after training, thus improving the performance of the algorithm. Second, in order to better adapt to the small target scale and enhance the ability of small target feature acquisition, the SPD-Conv module was integrated in the backbone network, replacing the original stepwise convolutional layer and pooling layer, so as to solve the problems of loss of fine-grained information and low efficiency of feature representation existing in the current convolutional neural network (CNN) structure. In the neck part, nearest-neighbor upsampling was replaced by the feature reassembly assembly of features operator CARAFE (content-aware reassembly of features), which enabled the model to aggregate contextual information in a larger perceptual field and enhanced the feature representation in the neck. Finally, validation experiments were conducted by comparing different algorithms under the same VisDrone-2021 dataset. The results of the ablation experiments show that the algorithms proposed in this thesis have improved the recall (R), mAP50, and mAP95 by 4.7, 5.8 and 5.7%, respectively, compared with YOLOv8n. The results of the model generalization experiments on the TinyPerson dataset show that the algorithm in this paper has superior small target detection performance with only 1.2 M model parameters (1 M = 106). [ABSTRACT FROM AUTHOR]

Cluster analysis is a principal approach to discover unknown tumor subtypes. Innovative and effective cluster analysis methods are of great significance for tumor diagnosis and malignant tumor treatment. Existing studies on the cluster analysis of tumor gene data generally have defects in aspects such as unsatisfactory performance in clustering highdimensional and high-noise data, and insufficient accuracy in selecting cluster centers. To overcome these defects, this paper performed cluster analysis on tumor gene data based on an improved Density peaks clustering (DPC) algorithm. At first, this paper elaborated on the composition and storage format of tumor tissue samples used in the experiment, gave the tumor gene expression profile data in the matrix format, and introduced the preprocessing process of gene expression profile data. Then, this paper carried out feature selection of tumor gene expression profile data. At last, this paper innovatively divided the target gene density into two parts of K-nearest neighbor local density and neighborhood density, thereby completing the improvement of conventional DPC algorithm and expanding its application scenarios. Combining with experiment, the clustering results of the algorithm before and after introducing the idea of Approximate Nearest Neighbor (ANN) were given, which had verified the effectiveness of the algorithm proposed in this paper. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) is a platform that manages daily life tasks to establish an interaction between things and humans. One of its applications, the smart office that uses the Internet to monitor electrical appliances and sensor data using an automation system, is presented in this study. Some of the limitations of the existing office automation system are an unfriendly user interface, lack of IoT technology, high cost, or restricted range of wireless transmission. Therefore, this paper presents the design and fabrication of an IoT-based office automation system with a user-friendly smartphone interface. Also, real-time data monitoring is conducted for the predictive maintenance of sensor nodes. This model uses an Arduino Mega 2560 Rev3 microcontroller connected to different appliances and sensors. The data collected from different sensors and appliances are sent to the cloud and accessible to the user on their smartphone despite their location. A sensor fault prediction model based on a machine learning algorithm is proposed in this paper, where the k-nearest neighbors model achieved better performance with 99.63% accuracy, 99.59% F1-score, and 99.67% recall. The performance of both models, i.e., k-nearest neighbors and naive Bayes, was evaluated using different performance metrics such as precision, recall, F1-score, and accuracy. It is a reliable, continuous, and stable automation system that provides safety and convenience to smart office employees and improves their work efficiency while saving resources. [ABSTRACT FROM AUTHOR]

In the previous research on traffic flow prediction models, most of the models mainly studied the time series of traffic flow, and the spatial correlation of traffic flow was not fully considered. To solve this problem, this paper proposes a method to predict the spatio-temporal characteristics of short-term traffic flow by combining the k-nearest neighbor algorithm and bidirectional long short-term memory network model. By selecting the real-time traffic flow data observed on high-speed roads in the United Kingdom, the K-nearest neighbor algorithm is used to spatially screen the station data to determine the points with high correlation and then input the BILSTM model for prediction. The experimental results show that compared with SVR, LSTM, GRU, KNN-LSTM, and CNN-LSTM models, the model proposed in this paper has better prediction accuracy, and its performance has been improved by 77%, 19%, 18%, 22%, and 13%, respectively. The proposed K-nearest neighbor-bidirectional long short-time memory model shows better prediction performance. [ABSTRACT FROM AUTHOR]

The optical properties of Mn4+-activated phosphors are heavily affected by the host and the related data are difficult to obtain, which makes quantitative design of novel phosphors with required wavelength particularly challenging. In the era of big data, computation is emerging as a vital complement to materials development. Compared with traditional calculation routes, machine learning (ML) models provide better solutions for the problems wherein understanding the underlying physics is complicated with sparse data. In this study, we developed six ML models based on a sparse and small data set to rapidly predict the peak emission wavelength of novel phosphors and use Recursive Feature Elimination (RFE) to select the best features. Among them, the K-nearest neighbors regression (KNN.r) algorithm using non-parametric regression technology can predict the value of the target variable based on the similarity between the target value and its spatial neighbors, which can tackle sparse data sets with no obvious linear relationship well and has the best prediction accuracy. These models were then used to predict 278 potential Mn4+-activated phosphor hosts extracted from the inorganic crystal structure database (ICSD), and 19 samples were selected to validate these methods. In order to overcome the lack of physical and chemical insight of KNN, a visual data analysis was carried out based on hierarchical clustering analysis (HCA), and the rule between the peak emission wavelength and the structure of host was proposed. Guided by the results of KNN and HCA, six samples were selected for the synthesis and characterization to verify our conclusions. The measured values of all these phosphors exhibit good agreement with the predicted results, suggesting the success of our methodology. Thus, the model established and the rule proposed in this paper provide a new pathway for the rapid discovery of novel phosphors. [ABSTRACT FROM AUTHOR]

Removal of duplicate data is a new way used to compress the data by removing duplicate copies of information. It ensures data management by efficiently reducing the storage space and maintaining the energy consumption. Duplication improves storage utilization with higher reliability. Due to abundant data generation through various sources need of this system increases as it ensures data management by efficiently reducing the duplicate copies of data. This technique results in making a system more optimized by calculating hash value of the files. The following paper aims to achieve the above goal by detecting and eliminating the duplicate data. This proposed system is a simple framework to use, provides ease of retrieval of data from storage and calculate hash value by using SHA (secure hash algorithm) and d hash (difference hash algorithm). Data in the form of text, images, audio and video can be examined in this proposed paper. This paper proposes new hash functions for indexing local image descriptors. These functions are first applied and evaluated as a range neighbor algorithm. We show that it obtains similar results as several state-of-the-art algorithms. In the context of near duplicate image retrieval, we integrated the proposed hash functions within a bag of words approach. Because most of the other methods use a k-means based vocabulary, they require an off-line learning stage and highest performance is obtained when the vocabulary is learned on the searched database. For application where images are often added or removed from the searched dataset, the learning stage must be repeated regularly in order to keep high recalls. We show that our hash functions in a bag of words approach has similar recalls as bag of words with k-means vocabulary learned on the searched dataset, but our method does not require any learning stage. It is thus very well adapted to near duplicate image retrieval applications where the dataset evolves regularly as there is no need to update the vocabulary to guarantee the best performance. [ABSTRACT FROM AUTHOR]

Sampling a nearest neighbor is often presented as a Hansen-Hurwitz or Horvitz-Thompson estimation process. This may not be the most informative viewpoint, and measuring the probability of selection is not necessary. The measurement of the nearest object as a "depth" over the selection area can be done by a sampling process, and the total estimated without knowing the polygon areas. The process is unbiased, quite general, and easy to understand. It can be extended to more than just the nearest object to a sample point and to many different polygon shapes. This paper is an extension, simplification and generalization of an earlier paper in this journal (Iles, K. 2009. "Nearest-tree" estimations--A discussion of their geometry, MCFNS 1(2), pp. 47-51.), but does not require a random orientation or weighting for the direction of measurement from the tree to the polygon border. [ABSTRACT FROM AUTHOR]

Classification of the patterns is a crucial structure of research and applications. Using fuzzy set theory, classifying the patterns has become of great interest because of its ability to understand the parameters. One of the problems observed in the fuzzification of an unknown pattern is that importance is given only to the known patterns but not to their features. In contrast, features of the patterns play an essential role when their respective patterns overlap. In this paper, an optimal fuzzy nearest neighbor model has been introduced in which a fuzzification process has been carried out for the unknown pattern using k nearest neighbor. With the help of the fuzzification process, the membership matrix has been formed. In this membership matrix, fuzzification has been carried out of the features of the unknown pattern. Classification results are verified on a completely llabelled Telugu vowel data set, and the accuracy is compared with the different models and the fuzzy k nearest neighbor algorithm. The proposed model gives 84.86% accuracy on 50% training data set and 89.35% accuracy on 80% training data set. The proposed classifier learns well enough with a small amount of training data, resulting in an efficient and faster approach. [ABSTRACT FROM AUTHOR]

For most bioinformatics statistical methods, particularly for gene expression data classification, prognosis, and prediction, a complete dataset is required. The gene sample value can be missing due to hardware failure, software failure, or manual mistakes. The missing data in gene expression research dramatically affects the analysis of the collected data. Consequently, this has become a critical problem that requires an efficient imputation algorithm to resolve the issue. This paper proposed a technique considering the local similarity structure that predicts the missing data using clustering and top K nearest neighbor approaches for imputing the missing value. A similarity-based spectral clustering approach is used that is combined with the K-means. The spectral clustering parameters, cluster size, and weighting factors are optimized, and after that, missing values are predicted. For imputing each cluster's missing value, the top K nearest neighbor approach utilizes the concept of weighted distance. The evaluation is carried out on numerous datasets from a variety of biological areas, with experimentally inserted missing values varying from 5 to 25%. Experimental results prove that the proposed imputation technique makes accurate predictions as compared to other imputation procedures. In this paper, for performing the imputation experiments, microarray gene expression datasets consisting of information of different cancers and tumors are considered. The main contribution of this research states that local similarity-based techniques can be used for imputation even when the dataset has varying dimensionality and characteristics. [ABSTRACT FROM AUTHOR]

This paper presents an effective machine learning-based depth selection algorithm for CTU (Coding Tree Unit) in HEVC (High Efficiency Video Coding). Existing machine learning methods are limited in their ability in handling the initial depth decision of CU (Coding Unit) and selecting the proper set of input features for the depth selection model. In this paper, we first propose a new classification approach for the initial division depth prediction. In particular, we study the correlation of the texture complexity, QPs (quantization parameters) and the depth decision of the CUs to forecast the original partition depth of the current CUs. Secondly, we further aim to determine the input features of the classifier by analysing the correlation between depth decision of the CUs, picture distortion and the bit-rate. Using the found relationships, we also study a decision method for the end partition depth of the current CUs using bit-rate and picture distortion as input. Finally, we formulate the depth division of the CUs as a binary classification problem and use the nearest neighbor classifier to conduct classification. Our proposed method can significantly improve the efficiency of interframe coding by circumventing the traversing cost of the division depth. It shows that the mentioned method can reduce the time spent by 34.56% compared to HM-16.9 while keeping the partition depth of the CUs correct. [ABSTRACT FROM AUTHOR]

Approximate nearest neighbor search (ANNS) for high dimensional data has received extensive research efforts. Many existing ANNS methods in metric spaces are essentially based on the permutation of pivots, or pre-selected reference points, which are ordered by their distances to the data. Unfortunately, most of these permutation-based methods are quite memory demanding, because of either complexity of index structure, excessive pivots, or insufficient exploitation of distances. This paper proposes the reduced permutation array (RPA), a delicate integration of existing technical elements based on in-depth understanding, for recall@R ANNS in metric spaces. As an array, for each data element, RPA stores only l identities (IDs) of its nearest pivots out of k pre-selected ones (l « k). During search, the order of distances from each data element to the query is approximated by a score function based on the distances from its nearest pivots to the query, and a bounded heap is maintained to store R candidate results. RPA is simple, memory-efficient, and scalable. Experiments show that RPA is approximately 3 times more memory-efficient than existing methods on average. The research results can provide an effective ANNS method for massive data in a single-machine environment with limited memory resources. [ABSTRACT FROM AUTHOR]

Time series classification models have been garnering significant importance in the research community. However, not much research has been done on generating adversarial samples for these models. These adversarial samples can become a security concern. In this paper, we propose utilizing an adversarial transformation network (ATN) on a distilled model to attack various time series classification models. The proposed attack on the classification model utilizes a distilled model as a surrogate that mimics the behavior of the attacked classical time series classification models. Our proposed methodology is applied onto 1-nearest neighbor dynamic time warping (1-NN DTW) and a fully convolutional network (FCN), all of which are trained on 42 University of California Riverside (UCR) datasets. In this paper, we show both models were susceptible to attacks on all 42 datasets. When compared to Fast Gradient Sign Method, the proposed attack generates a larger faction of successful adversarial black-box attacks. A simple defense mechanism is successfully devised to reduce the fraction of successful adversarial samples. Finally, we recommend future researchers that develop time series classification models to incorporating adversarial data samples into their training data sets to improve resilience on adversarial samples. [ABSTRACT FROM AUTHOR]

With more locations for wind generation, the grid's dependability is degraded. This paper presents a state-of-art combined wind power prediction model, including data preprocessing, improved secondary decomposition, and deep learning. A density-based spatial clustering of applications with noise was used primarily to identify and address irrational data and then correct them using k-nearest neighbor. Later, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) was used to decompose the original wind power time series into several intrinsic mode functions (IMFs), and the variational mode decomposition (VMD) was adopted for further decomposition, due to its high irregularity and instability, of the first two components. Finally, a long short-term memory (LSTM) was employed to predict each component. The proposed model was then applied to two wind farms in Turkey and France. The experimental findings are as follows: (1) The data preprocessing scheme proposed in this paper can improve the predicted results. After data preprocessing, mean absolute error (MAE) and root mean squared error (RMSE) have declined by 10.73% and 10.20% on average, respectively. (2) The improved predictions were greater than the common secondary decomposition. The MAE and RMSE of improved CEEMDAN-VMD-LSTM were down by 14.77% and 15.12% on average, compared with CEEMDAN-VMD-LSTM, respectively. [ABSTRACT FROM AUTHOR]

In recent years, there exist few difficulties for textile industries to predict the yarn nep index for small data and data with mutation. To fill this gap, a yarn nep prediction method combining grey correlation analysis and nearest-neighbour prediction method is proposed. In this paper, 26 indicators such as the raw cotton quality indicators and key process parameters are used as the input of the prediction model for yarn nep. The experimental results show that the relative error of the new method is lower than 10%, while the relative error of the individual data predicted by the traditional three-layer BP neural network is very large. Compared with the BP neural network, the average relative error and root-mean-square error of our proposed method are smaller, while the data are stable and the volatility is small. The prediction performance meets the user's requirements. The effectiveness of the proposed model is proved. [ABSTRACT FROM AUTHOR]

Internet of things (IoT) comprises many heterogeneous nodes that operate together to accomplish a human friendly or a business task to ease the life. Generally, IoT nodes are connected in wireless media and thus they are prone to jamming attacks. In the present scenario jamming detection (JD) by using machine learning (ML) algorithms grasp the attention of the researchers due to its virtuous outcome. In this research, jamming detection is modelled as a classification problem which uses several features. Using one/two or minimumnumber of features produces vague results that cannot be explained. Also the relationship between the feature and the class label cannot be efficiently determined, specifically, if the chosen number of features for training is minimum (say 1 or 2). To obtain good results, machine-learning algorithms are trained by large number of data sets. However, collection of large number of datasets to solve jamming detection is not easy and most of the times generation and collection of large data sets become paradigmatic. In this paper, to solve this problem, more number of features with nominal number of data's is considered that eases the data collection and the classification accuracy. In this research, an efficient technique based on locality sensitive hashing (LSH) for K-nearest neighbor algorithm (K-NN), which takes less time for constructing and querying the hash table that gives good accuracy is proposed and evaluated. From the results, it is clear that the obtained results are validatable and the model is more sensible. [ABSTRACT FROM AUTHOR]

Trajectory-based spatiotemporal entity linking is to match the same moving object in different datasets based on their movement traces. It is a fundamental step to support spatiotemporal data integration and analysis. In this paper, we study the problem of spatiotemporal entity linking using effective and concise signatures extracted from their trajectories. This linking problem is formalized as a $k$ k -nearest neighbor ($k$ k -NN) query on the signatures. Four representation strategies (sequential, temporal, spatial, and spatiotemporal) and two quantitative criteria (commonality and unicity) are investigated for signature construction. A simple yet effective dimension reduction strategy is developed together with a novel indexing structure called the WR-tree to speed up the search. A number of optimization methods are proposed to improve the accuracy and robustness of the linking. Our extensive experiments on real-world datasets verify the superiority of our approach over the state-of-the-art solutions in terms of both accuracy and efficiency. [ABSTRACT FROM AUTHOR]

In order to better understand the purchase decision-making process of consumers, this paper makes an in-depth study on the precision marketing of e-commerce products on the basis of KNN algorithm. Through data mining, classic KNN algorithm, BPNN algorithm, and other methods, this paper takes the price and purchase intention of e-commerce agricultural products as an example. Based on the classic nearest neighbor algorithm, binomial function is combined with Euclidean distance formula when calculating the nearest neighbor through similarity. The particle swarm optimization algorithm is used to optimize the binomial function coefficient and the K value of the nearest neighbor algorithm, and the results of the best prediction model for the prediction application of e-commerce agricultural product price and purchase intention are established. Both pricing strategies and promotion strategies will weaken the compromise effect of consumers when they choose e-commerce agricultural products. After studying the calculation method of the KNN algorithm, it not only correctly predicts the price of e-commerce agricultural products but also makes a corresponding prediction and analysis of consumers' purchase intention of e-commerce agricultural products, with the highest accuracy of 94.2%. At the same time, in the future precision marketing process, e-commerce agricultural products enterprises use data technology to achieve precision marketing, which effectively changes the shortcomings of traditional marketing and improves the product marketing effect and economic benefits. [ABSTRACT FROM AUTHOR]

In view of the problem of excessive noise in financial time series, this paper proposes a nearest neighbor dynamic time warping classification method for financial time series based on the optimal smooth denoising model (osdDTW2). First, the optimal smooth denoising model is improved, and then the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) method is used to decompose the time series signal. Then, the improved optimal smooth denoising model is used to construct a low-pass filter to do the denoising of the time series. After being denoised, the time series are aligned by dynamic time warping (DTW), Finally, the nearest neighbor method is used for classification. This paper also uses the UCR datasets to verify the effectiveness of the proposed method and applies it method to financial time series classification. The experimental results suggest that osdDTW2 ( α sm = 0. 9) can improve the effectivness of the benchmark algorithm (DTW) to some extent. [ABSTRACT FROM AUTHOR]

The aim of this paper is to select sites for health care centers in urban areas using GIS by suggesting a framework for selecting the best locations proposed for new health care centers, also focuses on access to health care centers, geographical distribution, and managing various health care (HC) facilities in accordance to the criteria and standards of the Ministry of Health and Population (MOHP). This paper relied on program ArcGIS 10.6 to determine the geographical distribution, determine the nearest neighbor, the intermediate location, the standard distance and Inclination, the buffer zone and the intersection, and analysis the road map to discover the nearest health care center. The proposed methodology involved many procedures, based on using satellite images, GIS, demographic, health data. The obtained results showed that there was shortage in PHC centers in the selected study area. As the research concluded that, there are many other governorate and private health service, which should be studied in details and from using this framework are finding the best locations for health care services centers by identifying the nearest access to health care centers and the geographical distribution of health care services. [ABSTRACT FROM AUTHOR]

In this paper, we propose an image completion algorithm based on dense correspondence between the input image and an exemplar image retrieved from the Internet. Contrary to traditional methods which register two images according to sparse correspondence, in this paper, we propose a hierarchical PatchMatch method that progressively estimates a dense correspondence, which is able to capture small deformations between images. The estimated dense correspondence has usually large occlusion areas that correspond to the regions to be completed. A nearest neighbor field (NNF) interpolation algorithm interpolates a smooth and accurate NNF over the occluded region. Given the calculated NNF, the correct image content from the exemplar image is transferred to the input image. Finally, as there could be a color difference between the completed content and the input image, a color correction algorithm is applied to remove the visual artifacts. Numerical results show that our proposed image completion method can achieve photo realistic image completion results. [ABSTRACT FROM AUTHOR]

It is essential to define more convincing and applicable classifiers for small datasets. In this paper, a minimum reconstruction error-based K-nearest neighbors (K-NN) classifier is proposed. We propose a new neighbor selection method. In the proposed neighbor selection method, a subset of data that minimize the reconstruction error is assigned as the neighbors of the query sample. Also, there is not any constraint on the distance of the neighbors from the query sample. An l 0 -based sparse representation problem is introduced for selecting the neighbors. These neighbors are assigned as neighbors of the minimum reconstruction error-based K-NN classifiers. Three l 0 -based minimum reconstruction error-based K-NN classifiers are introduced. These classifiers are less sensitive to the reconstruction coefficients in minimum reconstruction error-based K-NN classifiers and reconstruct the query sample with less error. The results on UCI machine learning repository, UCR time-series archive datasets, and a small subset (16%) of MNIST handwritten digit database demonstrate the suitable performance of the proposed method. The recognition precision increases up to more than 3% in some evaluations. [ABSTRACT FROM AUTHOR]

This paper introduces a novel modification to the Inverted File (IVF) index approach for approximate nearest neighbor search, incorporating supervised learning techniques to enhance the efficacy of intermediate clustering and achieve more balanced cluster sizes. The proposed method involves creating clusters using a neural network by solving a task to classify query vectors into the same bucket as their corresponding nearest neighbor vectors in the original dataset. When combined with minimizing the standard deviation of the bucket sizes, the indexing process becomes more efficient and accurate during the approximate nearest neighbor search. Through empirical evaluation on a test dataset, we demonstrate that the proposed semi-supervised IVF index approach outperforms the industry-standard IVF implementation with fixed parameters, including the total number of clusters and the number of clusters allocated to queries. This novel approach has promising implications for enhancing nearest-neighbor search efficiency in high-dimensional datasets across various applications, including information retrieval, natural language search, recommendation systems, etc. [ABSTRACT FROM AUTHOR]

In order to realize intelligent identification of rail damage, this paper studies the extraction method of complete damage ultrasonic B-scan data based on the density-based spatial clustering of applications with noise algorithm (DBSCAN). Aiming at the problem that the traditional DBSCAN algorithm needs to manually set the Eps and Minpts parameters, a KMNN-DBSCAN (K-median nearest neighbor DBSCAN) algorithm is proposed. The algorithm first uses the dataset's own distribution characteristics to generate a list of Eps and Minpts parameters and then determines the optimal Eps and Minpts through an optimization strategy to achieve complete self-adaptation of the two parameters of Eps and Minpts. In order to further improve the clustering performance of the algorithm, the partition idea is introduced, and the partition KMNN-DBSCAN algorithm is proposed to solve the problem that the clustering results of the DBSCAN algorithm are inconsistent with the actual categories on datasets with uneven density. The experimental results show that the KMNN-DBSCAN algorithm has higher clustering accuracy and silhouette coefficient (SC) for the D037 dataset ultrasound information group (UIG) division; compared with the KMNN-DBSCAN algorithm, the proposed partition KMNN-DBSCAN algorithm has higher clustering accuracy, F-Measure, and SC values. The partition KMNN-DBSCAN algorithm achieves accurate division of all damage UIG on the damaged B-scan data with large density differences, and completes the effective extraction of complete damage data. [ABSTRACT FROM AUTHOR]

The data gathering and composite the services through the IoT devices are the significant need in current scenario. There are many existing systems which gather the data from IoT devices and provide as an analysis based on the service linked with them. The objective of this paper is to build a middleware for IOT tech stack, which can recognize different services and features and categorize them via a ranking solution similar to the page ranking algorithm. The service ranking algorithm (SRA) are linked with HMM to fixates on domain specific requirements and controls services based on said fixated domain. Hence, this algorithm has to be dynamic in nature and should be able to accommodate different domain schemas as possible, where the parameters of distinction for each domain is to be pre specified and the algorithm is to be tuned accordingly. Before ranking begins, selecting the relevant service based on the availability of services, the Service Provider has to decide the kind of services to be offered for the clients based on the weight's reliability, completeness and energy availability. For implementing this, many intelligent systems are suggested to choose the low cost and high reliable services. In this chapter, a fuzzy rule-based K Nearest Neighbour Classifier is used for categorize the IoT service based on user request. In addition, XGBoost (Extreme Gradient Boosting) is used for dynamic service selection from the available categorized services based on response time and cost. Hidden Markov Model (HMM) is the prediction model used in this proposed work to solve the energy prediction problem. [ABSTRACT FROM AUTHOR]

The early hepatic fibrosis staging is very important for timely diagnosis, prognosis, and treatment of all chronic liver diseases. Diffraction-enhanced imaging, which can provide much more information on soft tissue morphology than conventional absorption radiography, might be a potential noninvasive technique to diagnose and stage hepatic fibrosis. This paper presents different feature extraction strategies and classification methods to automatically classify hepatic fibrosis using diffraction-enhanced imaging images. Texture features are obtained using a total of three methods including first order feature, gray level co-occurrence matrix, and grayscale gradient co-occurrence matrix. The fusion of these texture features is also studied. The principal component analysis is used to reduce the dimension of the features and redundant information among data. The features are classified using two popular classification techniques, namely, K-nearest neighbors and support vector machines. On the basis of the comparison of different feature strategies and classification methods, we can identify the suitable methods for grading hepatic fibrosis. The proposed approach efficiently classifies the hepatic fibrosis DEI images into four classes with the highest classification accuracy of 99.99%. We further demonstrate the potential of the DEI images in staging hepatic fibrosis. [ABSTRACT FROM AUTHOR]

In electronic markets, malicious sellers often employ reviewers to carry out different types of attacks to improve their own reputations or destroy their opponents' reputations. As such attacks may involve deception, collusion, and complex strategies, maintaining the robustness of reputation evaluation systems remains a challenging problem. From a platform manager's view, no trader can be taken as a trustable benchmark for reference, therefore, accurate filtration of dishonest sellers and fraud reviewers and precise presentation of users' reputations remains a challenging problem. Based on impression theory, this paper presents an unsupervised strategy, which first design a nearest neighbor search algorithm to select some typical lenient reviewers and strict reviewers. Then, based on these selected reviewers and the behavior expectation theory in impression theory, this paper adopts a classification algorithm that pre-classify sellers into honest and dishonest ones. Thirdly, another classification algorithm is designed to classify reviewers (i.e., buyers) into honest, dishonest, and uncertain ones according to their trading experiences with the pre-classified sellers. Finally, based on the ratings of various reviewers, this paper proposes a formula to estimate seller reputations. We further designed two general sets of experiments over simulated data and real data to evaluate our scheme, which demonstrate that our unsupervised scheme outperforms benchmark strategies in accurately estimating seller reputations. In particular, this strategy can robustly defend against various common attacks and unknown attacks. [ABSTRACT FROM AUTHOR]

In the past few decades, quantum computation has become increasingly attractive due to its remarkable performance. Quantum image scaling is considered a common geometric transformation in quantum image processing, however, the quantum floating-point data version of which does not exist. Is there a corresponding scaling for 2-D and 3-D floating-point data? The answer is yes. In this paper, we present a quantum scaling up and down scheme for floating-point data by using trilinear interpolation method in 3-D space. This scheme offers better performance (in terms of the precision of floating-point numbers) for realizing the quantum floating-point algorithms than previously classical approaches. The Converter module we proposed can solve the conversion of fixed-point numbers to floating-point numbers of arbitrary size data with p + q qubits based on IEEE-754 format, instead of 32-bit single-precision, 64-bit double-precision and 128-bit extended-precision. Usually, we use nearest-neighbor interpolation and bilinear interpolation to achieve quantum image scaling algorithms, which are not applicable in high-dimensional space. This paper proposes trilinear interpolation of floating-point data in 3-D space to achieve quantum algorithms of scaling up and down for 3-D floating-point data. Finally, the quantum scaling circuits of 3-D floating-point data are designed. [ABSTRACT FROM AUTHOR]

Approximate nearest neighbor (ANN) search has become an essential paradigm for large-scale image retrieval. Conventional ANN search requires the categories of query images to been seen in the training set. However, facing the rapid evolution of newly-emerging concepts on the web, it is too expensive to retrain the model via collecting labeled data with the new (unseen) concepts. Existing zero-shot hashing methods choose the semantic space or intermediate space as the embedding space, which ignore the inconsistency of visual space and semantic space and suffer from the hubness problem on the zero-shot image retrieval task. In this paper, we present an novel deep quantization network with visual-semantic alignment for efficient zero-shot image retrieval. Specifically, we adopt a multi-task architecture that is capable of 1) learning discriminative and polymeric image representations for facilitating the visual-semantic alignment; 2) learning discriminative semantic embeddings for knowledge transfer; and 3) learning compact binary codes for aligning the visual space and the semantic space. We compare the proposed method with several state-of-the-art methods on several benchmark datasets, and the experimental results validate the superiority of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, we consider a new method dealing with the problem of estimating the scoring function γ a , with a constant a , in functional space and an unknown scale parameter under a nonparametric robust regression model. Based on the k Nearest Neighbors ( k NN) method, the primary objective is to prove the asymptotic normality aspect in the case of a stationary ergodic process of this estimator. We begin by establishing the almost certain convergence of a conditional distribution estimator. Then, we derive the almost certain convergence (with rate) of the conditional median (scale parameter estimator) and the asymptotic normality of the robust regression function, even when the scale parameter is unknown. Finally, the simulation and real-world data results reveal the consistency and superiority of our theoretical analysis in which the performance of the k NN estimator is comparable to that of the well-known kernel estimator, and it outperforms a nonparametric series (spline) estimator when there are irrelevant regressors. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a brain disorder that causes speech and communication problems. The speech issues are often described as slow speech and difficulty with articulation because Jaw muscles don’t move with enough strength. Mainly, clinical experts make a voice assessment to analyze voice signals, to detect PD. In this paper, we designed an intelligent model based on discrete wavelet transform (DWT) for the detection of Parkinson's. Firstly, the voice signals are passed through DWT. The suggested DWT-based model is then used to extract a collection of statistical features from voice data. The extracted features are evaluated, and the significant ones are chosen. The features that have been chosen are fed into a least squares support vector machine (LS-SVM), as well as another four classifiers K-nearest neighbor (KNN), bagged tree, SVM (support vector machine), and k means for the comparison. The proposed model is evaluated using a publicly available dataset named UCI machine learning repository. Several tests were carried out, and the findings revealed that the proposed framework can classify voice signals with a100% accuracy in the LS-SVM technique. [ABSTRACT FROM AUTHOR]

Nearest neighbour similarity measures are widely used in many time series data analysis applications. They compute a measure of similarity between two time series. Most applications require tuning of these measures' meta-parameters in order to achieve good performance. However, most measures have at least O (L 2) complexity, making them computationally expensive and the process of learning their meta-parameters burdensome, requiring days even for datasets containing only a few thousand series. In this paper, we propose UltraFastMPSearch, a family of algorithms to learn the meta-parameters for different types of time series distance measures. These algorithms are significantly faster than the prior state of the art. Our algorithms build upon the state of the art, exploiting the properties of a new efficient exact algorithm which supports early abandoning and pruning for most time series distance measures. We show on 128 datasets from the UCR archive that our new family of algorithms are up to an order of magnitude faster than the previous state of the art. [ABSTRACT FROM AUTHOR]

This paper presents a study on automatic title generation for scientific articles considering sentence information types known as rhetorical categories. A title can be seen as a high-compression summary of a document. A rhetorical category is an information type conveyed by the author of a text for each textual unit, for example: background, method, or result of the research. The experiment in this study focused on extracting the research purpose and research method information for inclusion in a computer-generated title. Sentences are classified into rhetorical categories, after which these sentences are filtered using three methods. Three title candidates whose contents reflect the filtered sentences are then generated using a template-based or an adaptive K-nearest neighbor approach. The experiment was conducted using two different dataset domains: computational linguistics and chemistry. Our study obtained a 0.109-0.255 F1- measure score on average for computer-generated titles compared to original titles. In a human evaluation the automatically generated titles were deemed 'relatively acceptable' in the computational linguistics domain and 'not acceptable' in the chemistry domain. It can be concluded that rhetorical categories have unexplored potential to improve the performance of summarization tasks in general. [ABSTRACT FROM AUTHOR]

Inspecting surface conditions in 3-D environments such as steel bridges is a complex, time-consuming, and often hazardous undertaking that is an essential part of tasks such as bridge maintenance. Developing an autonomous exploration strategy for a mobile climbing robot would allow for such tasks to be completed more quickly and more safely than is possible with human inspectors. The exploration strategy tested in this paper, called the nearest neighbors exploration approach (NNEA), aims to reduce the overall exploration time by reducing the number of sensor position evaluations that need to be performed. NNEA achieves this by first considering at each time step only a small set of poses near to the current robot as candidates for the next best view. This approach is compared with another exploration strategy for similar robots performing the same task. The improvements between the new and previous strategy are demonstrated through trials on a test rig, and also in field trials on a ferromagnetic bridge structure. [ABSTRACT FROM AUTHOR]

In recent years, collaborative representation (CR)-based models have been widely used in hyperspectral image (HSI) classification. And many CR models utilize Tikhonov regularization term to improve the classification performance. However, this regularization term greatly increases the running time of CR models due to calculating the Euclidean distance between the test sample and all the training samples. Moreover, most of CR models utilize the training samples of all classes to represent the test sample, which may make the classes irrelevant to the test sample produce an impact on the classification performance of CR. To solve the above problems and further improve the classification performance of the CR model, a novel Euclidean distance-based adaptive (EDA) dictionary learning method is proposed in this paper, which aims to construct a subdictionary by selecting the k-nearest neighbor classes relevant to the test sample using Euclidean distance. And the EDA method is integrated into the original collaborative representation classifier (CRC) and the CRC with Tikhonov regularization (CRT) methods for HSI classification, which is denoted as EDACRC and EDACRT, respectively. The proposed methods are evaluated on the Botswana, KSC, and Salinas-A datasets, in which the classification performances and running times of these methods are analyzed and compared. The experimental results demonstrate that the proposed EDA method can not only effectively reduce the running time of the CR model with Tikhonov regularization, but also effectively eliminate some classes irrelevant to the test sample, thus improving the classification performance of the CR model. [ABSTRACT FROM AUTHOR]

Sensitivity analysis is popular in dealing with missing data problems particularly for non-ignorable missingness, where full-likelihood method cannot be adopted. It analyses how sensitively the conclusions (output) may depend on assumptions or parameters (input) about missing data, i.e. missing data mechanism. We call models with the problem of uncertainty sensitivity models. To make conventional sensitivity analysis more useful in practice we need to define some simple and interpretable statistical quantities to assess the sensitivity models and make evidence based analysis. We propose a novel approach in this paper on attempting to investigate the possibility of each missing data mechanism model assumption, by comparing the simulated datasets from various MNAR models with the observed data non-parametrically, using the K-nearest-neighbour distances. Some asymptotic theory has also been provided. A key step of this method is to plug in a plausibility evaluation system towards each sensitivity parameter, to select plausible values and reject unlikely values, instead of considering all proposed values of sensitivity parameters as in the conventional sensitivity analysis method. The method is generic and has been applied successfully to several specific models in this paper including meta-analysis model with publication bias, analysis of incomplete longitudinal data and mean estimation with non-ignorable missing data. [ABSTRACT FROM AUTHOR]

Green plant species identification plays an important role in so many aspects, such as ecological environment protection, Chinese medicine preparation, agricultural and horticultural application, etc. A method on green plants recognition based on wavelet transform and variable local edge patterns (VLEP) is proposed in this paper. Firstly, the original image is decomposed by wavelet transformation. Then texture features are extracted using VLEPs. At the same time, block-based and multi-resolution ideas are considered together to extract features after images are transformed by wavelet. Finally, the fused texture features are classified by the nearest neighbor method. The experimental results show that the proposed method is a promising method for recognizing the common green plants with the natural and complex background compared with the other state-of-the-art methods, and combination of block-based and multi-resolution ideas can further improve the accuracy rate effectively. [ABSTRACT FROM AUTHOR]

Time series classification is a fundamental problem in the time series mining community. Recently, many sophisticated methods which can produce state-of-the-art classification accuracy on the UCR archive have been proposed. Unfortunately, most of them are parameter-laden methods and require fine-tune for different datasets. Besides, training these classifiers is very computationally demanding, which makes them difficult to use in many real-time applications and previously unseen datasets. In this paper, we propose a novel parameter-light algorithm, MDTW, to classify time series. MDTW has a few parameters which do not require any fine-tune and can be chosen arbitrarily because the classification accuracy is largely insensitive to the parameters. MDTW has no training step; thus, it can be directly applied to unseen datasets. MDTW is based on a popular method, namely the nearest neighbor classifier with Dynamic Time Warping (NN-DTW). However, MDTW performs much faster than NN-DTW by representing time series in different resolutions and using filters-and-refine framework to find the nearest neighbor. The experimental results demonstrate that MDTW performs faster than the state-of-the-art, with small losses (<3%) in average classification accuracy. Besides, we embed a technique, prunedDTW, into the MDTW procedure to make MDTW even faster, and show by experiments that this combination can speed up the MDTW from one to five times. [ABSTRACT FROM AUTHOR]

Data-driven preventive scanning for transient stability assessment (DTSA) is a faster and more efficient solution than time-domain simulation (TDS). However, most current methods cannot balance generalization to different topologies and interpretability, with simple output. A model that conforms to the physical mechanism and richer label for transient stability can increase confidence in DTSA. Thus a static-information, k-neighbor, and self-attention aggregated schema (SKETCH) is proposed in this paper. Taking only static measurements as input, SKETCH gives several explanations that are consistent with the physical mechanisms of TSA and provides results for all generator stability while predicting system stability. A module based on the self-attention mechanism is designed to solve the locality problem of a graph neural network (GNN), achieving subgraph equivalence outside the k-order neighborhood. Test results on the IEEE 39-bus system and IEEE 300-bus system indicate the superiority of SKETCH and also demonstrate the rich sample interpretation results. Highlights: A fast TSA scheme for pre-failure scanning. A physical mechanism-based attention structure for dynamic graph pooling. A node regression model that responds to key physical mechanisms. Generator label for richer output information. Top performance and post-hoc interpretation. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The nearest neighbor model has been a popular model for spatial data. This model assumes a distance‐based neighborhood structure among a set of entities that define the observations in a dataset. Values of a dependent variable are a function of the corresponding values of its neighbors and values of its own explanatory variables. In this paper, we extend the concept of "spatial neighbors" to a more general network dependency where the nearest neighbor is not necessarily distance‐based. As each entity can belong to multiple groups, observations can be related by multiple networks. By exploring the overlapping information that are exhibited by an entity and its neighbors, we modify the traditional spatial autoregressive model with autoregressive disturbances (SARAR) in three ways. First, multiple networks are allowed. Second, differences between the explanatory variables and their corresponding neighboring averages are applied. Third, apart from independent innovations, we investigate the case where the innovations are also multiple‐network dependent. Statistical inference of the model parameters is achieved via maximizing the profile log‐likelihood function. The standard errors of the profile maximum likelihood estimators are calculated by the inverse Fisher information matrix. Four variations of the multiple network autoregressive models are applied to predict changes in stock prices of 555 companies listed in the Hong Kong Stock Exchange. The model with attribute differencing and dependent innovations produces a smaller out‐of‐sample mean prediction error sum of squares than that of the other models. [ABSTRACT FROM AUTHOR]

Density peak clustering is the latest classic density-based clustering algorithm, which can directly find the cluster center without iteration. The algorithm needs to determine a unique parameter, so the selection of parameters is particularly important. However, for multi-density data, when one parameter cannot satisfy all data, clustering often cannot achieve good results. Moreover, the subjective selection of cluster centers through decision diagrams is often not very convincing, and there are also certain errors. In view of the above problems, in order to achieve better clustering of multi-density data, this paper improves the density peak clustering algorithm. Aiming at the selection of parameter dc, the K-nearest neighbor idea is used to sort the neighbor distance of each data, draw a line graph of the K-nearest neighbor distance, and find the global bifurcation point to divide the data with different densities. Aiming at the selection of cluster centers, the local density and distance of each data point in each data division is found, a γ map is drawn, the average value of the γ height difference is calculated, and through two screenings the largest discontinuity point is found to automatically determine the cluster center and the number of cluster centers. The divided datasets are clustered by the DPC algorithm, and then the clustering results are perfected and integrated by using the cluster fusion rules. Finally, a variety of experiments are designed from various perspectives on various artificial simulated datasets and UCI real datasets, which demonstrate the superiority of the F-DPC algorithm in terms of clustering effect, clustering quality, and number of samples. [ABSTRACT FROM AUTHOR]

Barrier coverage is a fundamental problem in wireless sensor networks, which can be applied to detect intruders that attempt to cross the boundaries of the region we are supervising. Among a number of issues related to barrier coverage, the most critical one is how to deploy sensors to construct barriers. Many barrier-building approaches have been proposed in previous papers by other scholars, while few papers adopt distributed strategies which allow sensors to move autonomously based on local information and are more practical in some unreachable areas. In this paper, we study a decentralized approach based on the nearest neighbor rule (NNR) to realize the self-deployment of wireless sensor networks to build K strong barriers, while considering network connectivity. This approach can be divided into three stages which correspond to the three times of using the NNR respectively. The sensor can autonomously choose the closest barrier to belonging to, which ensures the total migration distance is relatively smallest. Moreover, we test the availability and performance of our approach with simulations. [ABSTRACT FROM AUTHOR]

The paper gives insight on how the text document categorization problem can be solved and implemented in a software product. On that score, it specifies how input data are provided, processed and transformed into output data. The goal of the paper is not only to suggest a simple theoretical solution to the text document categorization problem but to provide a real-life implementation as part of a software system. [ABSTRACT FROM AUTHOR]

Background Breast cancer is the most common disease in women, and it is very difficult to detect breast cancer early, so there are fewer reasons to measure women's survival. The purpose of this paper to apply various basic machine learning algorithms on multiple breast cancer datasets and the obtained results combined with an ensemble technique to find whether the datasets getting high accuracy. Methods Five machine learning algorithms such as decision tree, random forest, k-nearest neighbor, support vector clustering (SVC) and logistic regression (LR) are used to achieve high accuracy. In this study, along with five classifiers and voting ensemble methods on three different datasets (Mammographic mass data, Original Wisconsin breast cancer data and Diagnostic Wisconsin breast cancer data) to obtain the predicted accuracy recall, precision and receiver operating characteristic (ROC) curves. The results obtained by three different datasets were compared. Results By applying five classifiers and voting ensemble, the comparison results show that logistic regression, SVC and voting ensemble methods are potentially effective for predicting classification accuracy. The highest value is (Accuracy: 97.13% with ensemble by D2, Precision: 96.90% with ensemble and LR method by D3, Recall: 97.78% with SVC by D3, ROC: 99.93% with LR by D2). Conclusions Machine learning classifiers and ensemble technologies have the potential to identify diseases. The applied method highlights the importance of classifiers and ensemble methods. [ABSTRACT FROM AUTHOR]

In this paper, some asymptotic results such as weak consistency, strong consistency, uniformly strong consistency and the corresponding rates for the nearest neighbour estimator of the density function are established under α-mixing samples. The results improve and extend some previous ones in the literature. As applications, the corresponding results on the consistencies for the estimator of the failure rate function in reliability theory are presented. [ABSTRACT FROM AUTHOR]

In the field of time series data mining, the accuracy of the simple, but very successful nearest neighbor (NN) classifier directly depends on the chosen similarity measure. To improve the efficiency of elastic measures introduced to overcome the shortcomings of Euclidean distance, the Sakoe-Chiba band is usually applied as a constraint. In this paper, we provide a detailed analysis of the influence of the alternative Itakura parallelogram constraint on the accuracy of the NN classifier in combination with four well-known elastic measures, compared to the Sakoe-Chiba constraint and the unconstrained variants of these measures. The findings suggest that, although the Sakoe-Chiba band generally produces better results, for certain types of datasets the Itakura parallelogram represents a better choice. [ABSTRACT FROM AUTHOR]

In an era characterized by fast technological progress that introduces new unpredictable scenarios every day, working in the law field may appear very difficult, if not supported by the right tools. In this respect, some systems based on Artificial Intelligence methods have been proposed in the literature, to support several tasks in the legal sector. Following this line of research, in this paper we propose a novel method, called PRILJ, that identifies paragraph regularities in legal case judgments, to support legal experts during the redaction of legal documents. Methodologically, PRILJ adopts a two-step approach that first groups documents into clusters, according to their semantic content, and then identifies regularities in the paragraphs for each cluster. Embedding-based methods are adopted to properly represent documents and paragraphs into a semantic numerical feature space, and an Approximated Nearest Neighbor Search method is adopted to efficiently retrieve the most similar paragraphs with respect to the paragraphs of a document under preparation. Our extensive experimental evaluation, performed on a real-world dataset provided by EUR-Lex, proves the effectiveness and the efficiency of the proposed method. In particular, its ability of modeling different topics of legal documents, as well as of capturing the semantics of the textual content, appear very beneficial for the considered task, and make PRILJ very robust to the possible presence of noise in the data. [ABSTRACT FROM AUTHOR]