Nearest-neighbor Algorithm Research Articles

Utilizing Nearest-Neighbor Clustering for Addressing Imbalanced Datasets in Bioengineering.

Imbalance classification is common in scenarios like fault diagnosis, intrusion detection, and medical diagnosis, where obtaining abnormal data is difficult. This article addresses a one-class problem, implementing and refining the One-Class Nearest-Neighbor (OCNN) algorithm. The original inter-quartile range mechanism is replaced with the K-means with outlier removal (KMOR) algorithm for efficient outlier identification in the target class. Parameters are optimized by treating these outliers as non-target-class samples. A new algorithm, the Location-based Nearest-Neighbor (LBNN) algorithm, clusters one-class training data using KMOR and calculates the farthest distance and percentile for each test data point to determine if it belongs to the target class. Experiments cover parameter studies, validation on eight standard imbalanced datasets from KEEL, and three applications on real medical imbalanced datasets. Results show superior performance in precision, recall, and G-means compared to traditional classification models, making it effective for handling imbalanced data challenges.

Improvement of the Nearest Neighbor Heuristic Search Algorithm for Traveling Salesman Problem

The Traveling Salesman Problem (TSP) is classified as a non-deterministic polynomial (NP) hard problem, which has found widespread application in several scientific and technological domains. Due to its NP-hard nature, it is very hard to solve effectively and efficiently. Despite this rationale, a multitude of optimization approaches have been proposed and developed by scientists and researchers during the last several decades. Among these several algorithms, heuristic approaches are deemed appropriate for addressing this intricate issue. One of the simplest and most easily implementable heuristic algorithms for TSP is the nearest neighbor algorithm (NNA). However, its solution quality suffers owing to randomness in the optimization process. To address this issue, this study proposes a deterministic NNA for solving symmetric TSP. It is an improved version of NNA, which starts with the shortest edge consisting of two cities and then repeatedly includes the closest city on the route until an effective route is established. The simulation is conducted on 20 benchmark symmetric TSP datasets obtained from TSPLIB. The simulation results provide evidence that the improved NNA outperforms the basic NNA throughout most of the datasets in terms of solution quality as well as computational time.

A heuristic‐based hybrid sampling method using a combination of SMOTE and ENN for imbalanced health data

AbstractHealth datasets often exhibit class imbalance, with healthy individuals being the majority class and patients being the minority class. As in many applications, machine learning methods are frequently used in the prediction and detection of diseases. The class imbalance presents a significant challenge for machine learning classifiers, particularly in their ability to accurately and effectively classify data from minority and majority classes. Therefore, data preprocessing is crucial before classifying the imbalanced data. In this study, we present GASMOTEPSO_ENN method that combines the synthetic minority oversampling technique (SMOTE) and edited nearest neighbour (ENN) algorithms using genetic algorithm (GA) and particle swarm optimization (PSO) heuristics as a preprocessing method to classify the imbalanced health datasets. In the experiments, chronic kidney disease (CKD), cerebral stroke prediction (CSP), and PIMA Indian diabetes (PID) datasets were utilized to assess the performance of the proposed method with metrics derived from the confusion matrix. The GASMOTEPSO_ENN method can classify the various diseases into different two classes of patients and healthy individuals with acceptable Matthews correlation coefficient (MCC) metric using the machine learning algorithms (Logistic regression (LR) 1.00 for CKD dataset, extreme gradient boosting (XGBoost) 0.94 for CSP dataset, and support vector machine (SVM) 0.87 for PID dataset). Moreover, the proposed method also performed well with other metrics in all datasets, and the analysis of the model results in relation to existing literature reveals that the proposed model demonstrably produces superior results.

A post-processing framework for class-imbalanced learning in a transductive setting

Traditional classification tasks suffer from the class-imbalanced problem, where some classes far outnumber others. To address this issue, existing class-imbalanced learning (CIL) methods either preprocess class-imbalanced datasets or adapt traditional classification algorithms to the imbalanced class distribution. Inspired by the idea of transductive learning, we propose a post-processing framework called PPF for CIL. Distinct from existing CIL methods, PPF directly adjusts the predicted labels of test data to fit the imbalanced class distribution. Specifically, we relabel some test data according to their prediction probabilities so that the class proportion of test data is close to that of training data. The underlying assumption is that training and test data, drawn independently from one data space, should obey the same class distribution. Furthermore, we propose a Compact Prototype-based Nearest Neighbor (CPNN) algorithm to assist the original classifier with the adjustment. Instead of training a classifier, CPNN classifies test data according to their distances to a set of prototypes estimated on labeled data. Thus, it is computationally simple and relatively robust to class imbalance. As a general framework, PPF can be easily applied to both traditional classification and CIL algorithms. To validate the effectiveness of the proposed method, we conducted extensive experiments on a variety of class-imbalanced datasets, using SVM and C4.5 as the original classifiers, respectively. Measured by F-measure, G-mean, and AUC, both PPF-SVM and PPF-C4.5 outperform 10 state-of-the-art CIL algorithms. Additionally, PPF further improved their performances when applied to 10 CIL algorithms.

Simplified Indoor Localization Using Bluetooth Beacons and Received Signal Strength Fingerprinting with Smartwatch.

Variations in Global Positioning Systems (GPSs) have been used for tracking users' locations. However, when location tracking is needed for an indoor space, such as a house or building, then an alternative means of precise position tracking may be required because GPS signals can be severely attenuated or completely blocked. In our approach to indoor positioning, we developed an indoor localization system that minimizes the amount of effort and cost needed by the end user to put the system to use. This indoor localization system detects the user's room-level location within a house or indoor space in which the system has been installed. We combine the use of Bluetooth Low Energy beacons and a smartwatch Bluetooth scanner to determine which room the user is located in. Our system has been developed specifically to create a low-complexity localization system using the Nearest Neighbor algorithm and a moving average filter to improve results. We evaluated our system across a household under two different operating conditions: first, using three rooms in the house, and then using five rooms. The system was able to achieve an overall accuracy of 85.9% when testing in three rooms and 92.106% across five rooms. Accuracy also varied by region, with most of the regions performing above 96% accuracy, and most false-positive incidents occurring within transitory areas between regions. By reducing the amount of processing used by our approach, the end-user is able to use other applications and services on the smartwatch concurrently.

Design and implementation of a Li River water quality monitoring and analysis system based on outlier data analysis.

The detection of water quality indicators such as Temperature, pH, Turbidity, Conductivity, and TDS involves five national standard methods. Chemically based measurement techniques may generate liquid residue, causing secondary pollution. The water quality monitoring and data analysis system can effectively address the issues that conventional methods require multiple pieces of equipment and repeated measurements. This paper analyzes the distribution characteristics of the historical data from five sensors at a specific time, displays them graphically in real time, and provides an early warning of exceeding the standard; It selects four water samples from different sections of the Li River, based on the national standard method, the average measurement errors of Temperature, PH, TDS, Conductivity and Turbidity are 0.98%, 2.23%, 2.92%, 3.05% and 3.98%.;It further uses the quartile method to analyze the outlier data over 100,000 records and five historical periods are selected. Experiment results show the system is relatively stable in measuring Temperature, PH and TDS, and the proportion of outlier is 0.42%, 0.84% and 1.24%. When Turbidity and Conductivity are measured, the proportion is 3.11% and 2.92%. In the experiment of using 7 methods to fill outlier, K nearest neighbor algorithm is better than others. The analysis of data trends, outliers, means, and extreme values assists in making decisions, such as updating and maintaining equipment, addressing extreme water quality situations, and enhancing regional water quality oversight.

A model reduction method for parametric dynamical systems defined on complex geometries

Dynamic mode decomposition (DMD) describes the dynamical system in an equation-free manner and can be used for the prediction and control. It is an efficient data-driven method for the complex systems. In this paper, we extend DMD to the parameterized problems and propose a model reduction method based on DMD to improve the computation efficiency. This method is an offline-online mechanism. In the offline phase, we need to generate the snapshots data by solving the parameterized equations for each parameter in the training set and perform the singular value decomposition (SVD) to get the reduced operator matrices, which would lead to the substantial computation. The weighted & interpolated nearest-neighbors algorithm (wiNN) is adopted to construct the efficient surrogate models of the reduced operator matrices (including the reduced Koopman operator matrix and SVD-modes). In the online phase, for each parameter, we only need to perform the operations based on the low-dimensional matrices to get the parameter DMD solution. Moreover, we choose the least squares radial basis function finite difference method for the spatial discretization. This can make our method more applicable to the parameterized problems defined on complex geometries. At last, the reaction-diffusion, the incompressible miscible flooding and the incompressible Navier-Stokes models defined on the complex geometries are presented to illustrate the effectiveness of the proposed method.

Exploring logistics transport route optimization: An algorithmic study based on RFID technology

With the rapid growth of e-commerce, logistics companies face challenges in efficient routing and scheduling to meet dynamic delivery demands. This paper proposes a novel logistics scheduling model to optimize vehicle routing using Radio Frequency Identification (RFID) technology. A vehicle scheduling model is developed. The random customer demand and service time are solved using an adaptive taboo search algorithm combined with a nearest neighbor algorithm. Comparative experiments demonstrate the performance of the improved method in completing tasks and reducing queueing time compared to other methods. A case study of route optimization for a logistics company shows the model can recommend optimized routes that reduce total transportation cost by over 25% compared to using RFID alone. The results highlight the potential of the proposed technique to enhance logistics efficiency. Limitations and future work are discussed.

Green Symphony: A Brief Review of Waste Segregation Techniques

Proper waste segregation poses challenges like insufficient awareness, inadequate infrastructure, and limited resources. In this study, we've scrutinized five waste segregation methods employing IoT, Arduino, Deep Learning, Machine Learning, and Artificial Intelligence. The IoT-based system utilizes sensors, and cameras to identify diverse waste types, transmit data to the IoT module, and activate sorting mechanisms like robotic arms [1]. This leads to remote monitoring, and automated waste sorting, fostering efficient recycling. An advanced garbage monitoring system with sensors and an Arduino Uno detects waste types, monitors garbage levels, and issues alerts when bins are full, contributing to waste management efficiency [2]. DL is integrated into the system, categorizing waste images using CNNs, and employing transfer nearest neighbor algorithm predicts waste management alerts based on three sensors, automating notifications to authorities with a 93.3% accuracy rate [3]. AI-driven waste segregation models, particularly for non-biodegradable plastics, reduce waste management costs, achieving over 80% accuracy in sorting plastics. The culmination is the (WSD) model [5], incorporating IoT, Arduino, ML, DL, and AI. Utilizing sensors and imaging devices, this model generates real-time data for efficient monitoring and sorting, featuring a K-nearest neighbor algorithm with 93.3% accuracy and DL precision in material classification. The WSD model provides an automated and eco-friendly solution, reducing manual labor and improving recycling practices for effective waste management. Advanced technology-based techniques like the WSD model offer solutions, emphasizing the importance of choosing technologies aligned with available resources and infrastructure for optimal impact.

Oversampling method via adaptive double weights and Gaussian kernel function for the transformation of unbalanced data in risk assessment of cardiovascular disease

In risk assessment of cardiovascular disease (CVD), the classification error caused by unbalanced data is a significant challenge, which has sparked widespread concern and research upsurge in the field of data mining. Therefore, in view of the imbalance of CVD data sets, an oversampling method via adaptive double weights and Gaussian kernel function (ADWGKFO) is proposed, which converts the unbalanced data sets into balanced data sets. Firstly, clustering algorithm is utilized to cluster minority samples, boundary samples are identified by Borderline-Synthetic Minority Over-sampling Technique (Borderline-SMOTE), K nearest neighbor and support vector machine algorithms, and the number of samples synthesized in each group is calculated based on the double weights of boundary points and majority distribution. Secondly, in order to clearly define the classification boundary, the mutual class potential of new samples in each cluster is calculated by Gaussian kernel function, and new samples are filtered according to the mutual class potential until the data set is balanced. Finally, taking the data sets from Kaggle platform as the research samples, the proposed method is empirically analyzed. In order to validate the efficacy and universality of the proposed method, this paper selects CatBoost that is a new integrated algorithm to test the effect of the ADWGKFO method, and compares it with different sampling methods and different classifiers using performance evaluation indexes such as accuracy, F1-score and area under the curve (AUC). Compared with the combinations of other methods, the accuracy, F1-score and AUC are significantly improved. It is concluded that the ADWGKFO method described in this paper can successfully improve the data quality, and increases the reliability of CVD risk assessment.

Absence of radio-bright dominance in a near-infrared selected sample of red quasars

Context. The dichotomy between red and blue quasars is still an open question. It is debated whether red quasars are simply blue quasars that are observed at certain inclination angles or if they provide insight into a transitional phase in the evolution of quasars. Aims. We investigate the relation between quasar colors and radio-detected fraction because radio observations of quasars provide a powerful tool in distinguishing between quasar models. Methods. We present the eHAQ+GAIA23 sample, which contains quasars from the High A(V) Quasar (HAQ) Survey, the Extended High A(V) Quasar (eHAQ) Survey, and the Gaia quasar survey. All quasars in this sample have been found using a near-infrared color selection of target candidates that have otherwise been missed by the Sloan Digital Sky Survey (SDSS). We implemented a redshift-dependent color cut in g* − i* to select red quasars in the sample and divided them into redshift bins, while using a nearest-neighbors algorithm to control for luminosity and redshift differences between our red quasar sample and a selected blue sample from the SDSS. Within each bin, we cross-matched the quasars to the Faint Images of the Radio Sky at Twenty centimeters (FIRST) survey and determined the radio-detection fraction. Results. For redshifts 0.8 < z ≤ 1.5, the red and blue quasars have a radio-detection fraction of 0.153−0.032+0.037 and 0.132−0.030+0.034, respectively. The red and blue quasars with redshifts 1.5 < z ≤ 2.4 have radio-detection fractions of 0.059−0.016+0.019 and 0.060−0.016+0.019, respectively, and the red and blue quasars with redshifts z > 2.4 have radio-detection fractions of 0.029−0.012+0.017 and 0.058−0.019+0.024, respectively. For the WISE color-selected red quasars, we find a radio-detection fraction of 0.160−0.034+0.038 for redshifts 0.8 < z ≤ 1.5, 0.063−0.017+0.020 for redshifts 1.5 < z ≤ 2.4, and 0.051−0.022+0.030 for redshifts z > 2.4. In other words, we find similar radio-detection fractions for red and blue quasars within < 1σ uncertainty, independent of redshift. This disagrees with what has been found in the literature for red quasars in SDSS. It should be noted that the fraction of broad absorption line (BAL) quasars in red SDSS quasars is about five times lower. BAL quasars have been observed to be more frequently radio quiet than other quasars, therefore the difference in BAL fractions could explain the difference in radio-detection fraction. Conclusions. The dusty torus of a quasar is transparent to radio emission. When we do not observe a difference between red and blue quasars, it leads us to argue that orientation is the main cause of quasar redness. Moreover, the observed higher proportion of BAL quasars in our dataset relative to the SDSS sample, along with the higher rate of radio detections, indicates an association of the redness of quasars and the inherent BAL fraction within the overall quasar population. This correlation suggests that the redness of quasars is intertwined with the inherent occurrence of BAL quasars within the entire population of quasars. In other words, the question why some quasars appear red or exhibit BAL characteristics might not be isolated; it could be directly related to the overall prevalence of BAL quasars in the quasar population. This finding highlights the need to explore the underlying factors contributing to both the redness and the frequency of BAL quasars, as they appear to be interconnected phenomena.

Su Altı Görüntü Sınıflandırma için HOG Özellik Çıkarıcı ve KNN Tabanlı Bir Yöntem

Underwater garbage affects the life of marine creatures and the entire ecosystem. Detecting underwater garbage is an important research area. In this study, a method is proposed to detect underwater garbage. The open-access Trash-ICRA19 dataset was used to implement the proposed method. The data set cropping process was applied and a data set consisting of 11060 images in total was obtained. These images were converted to 200×200 pixels using preprocessing. By applying the Directed Gradient Histogram (HOG) algorithm, 11060×900 feature vectors were obtained. The resulting feature vectors were then calculated using KNN (K Nearest Neighbor Algorithm), DT (Decision Tree), LD (Linear Discriminant), NB (Naive Bayes), and SVM (Support Vector Machine) classifiers. The results obtained showed that 97.78% accuracy was obtained when the KNN classifier was used in this method. The use of only feature extractors and classifiers in the proposed method shows that the method is lightweight. It has low computational complexity compared to existing studies in the literature. Moreover, according to its performance results, it is more successful than the methods in the literature.

Postoperative outcomes of transperitoneal versus retroperitoneal robotic partial nephrectomy: a propensity-score matched comparison focused on patient mobilization, return to bowel function, and pain

Literature meta-analyses comparing transperitoneal versus retroperitoneal approach to robotic partial nephrectomy (RPN) suggested some advantages favoring retroperitoneoscopy. Unfortunately, patient-centered data about mobilization, canalization, pain, and use of painkillers remained anecdotally reported. The present analysis aimed to compare transperitoneal versus retroperitoneal RPN focusing on such outcomes. Study data including baseline variables, perioperative, and postoperative outcomes of interest were retrieved from prospectively maintained institutional database (Jan 2018–May 2023) and compared between treatment groups (transperitoneal versus retroperitoneal). Propensity score matching was performed using the STATA command psmatch2 considering age, sex, body mass index, previous abdominal surgery, RENAL score, tumor size and location, and cT stage. The logit of propensity score was used for matching, with a 1:1 nearest neighbor algorithm, without replacement (caliper of 0.001). A total of 442 patients were included in the unmatched analysis: 330 underwent transperitoneal RPN 112 retroperitoneal RPN. After propensity score, 98 patients who underwent retroperitoneal RPN were matched with 98 patients who underwent transperitoneal RPN. Matched cohorts had comparable patients’ demographics and tumor features. We found similarity between the two laparoscopic accesses in all outcomes but in blood loss, which favored retroperitoneoscopic RPN (median 150 (IQR 100–300) versus 100 (IQR 0–100) ml, p = 0.03). No differences were found in terms of time to mobilization with ambulation, return to complete bowel function, postoperative pain, but higher painkillers consumption was reported after transperitoneal RPN (p < 0.004). The present study compared the transperitoneal versus the retroperitoneal approach to RPN, confirming the similarity between the two approaches in all perioperative outcomes. Based on our findings, the choice of the surgical approach to RPN may remain something that the surgeon decides.

Attacking Click-through Rate Predictors via Generating Realistic Fake Samples

How to construct imperceptible (realistic) fake samples is critical in adversarial attacks. Due to the sample feature diversity of a recommender system (containing both discrete and continuous features), traditional gradient-based adversarial attack methods may fail to construct realistic fake samples. Meanwhile, most recommendation models adopt click-through rate (CTR) predictors, which usually utilize black-box deep models with discrete features as input. Thus, how to efficiently construct realistic fake samples for black-box recommender systems is still full of challenges. In this article, we propose a hierarchical adversarial attack method against black-box CTR models via generating realistic fake samples, named CTRAttack. To better train the generation network, the weights of its embedding layer are shared with those of the substitute model, with both the similarity loss and classification loss used to update the generation network. To ensure that the discrete features of the generated fake samples are all real, we first adopt the similarity loss to ensure that the distribution of the generated perturbed samples is sufficiently close to the distribution of the real features, and then the nearest neighbor algorithm is used to retrieve the most appropriate features for non-existent discrete features from the candidate instance set. Extensive experiments demonstrate that CTRAttack can not only effectively attack the black-box recommender systems but also improve the robustness of these models while maintaining prediction accuracy.

The application and analysis of the KNN algorithm in machine learning for breast cancer prediction

When female patients are tested for cancer, breast cancer tends to be detected more often than other cancers. Breast cancer also accounts for a large percentage of deaths of female patients due to cancer. Timely prediction of whether one has breast cancer or not is of great significance in the treatment of breast cancer. In this study, K Nearest Neighbors (KNN) algorithm is used for data analysis as well as probability prediction to determine the probability of having breast cancer and compare it with traditional machine learning algorithms. The variation of accuracy versus K-value for a given dataset was experimentally represented by a graph. Also, the experiment gives and compares the accuracy of different learning algorithms in the same situation. According to the experimental results, the KNN algorithm tends to be more accurate than the other algorithms, with an accuracy of 98 percent. This turned out to be significantly higher than the remaining two algorithms. This experiment has significant implications for the healthcare industry. Because the use of this method can greatly improve the efficiency of doctors and the diagnosis of breast cancer, so that more people with the risk of breast cancer can be diagnosed and treated in a timely manner. In addition, hospitals can use the study to predict the incidence of breast cancer in a given area of the population.

Optimizing maintenance responsibility distribution in real estate management: A complexity-driven approach for sustainable efficiency

Efficient route management is critical for optimizing maintenance activities in real estate management. This study delves into the intricate task of allocating maintenance duties among building surveyors, a pivotal concern for property management firms. The primary goal of route planning is to enhance operational efficiency by minimizing travel time. To achieve this, the study explores three distinct algorithms: Saving, Nearest Neighbor, and an Evolutionary Algorithm (EA) customized with Dual Response Surface Optimization (DRSO). The integration of DRSO and EA enhances adaptability, allowing for dynamic responses to changes and improved allocation of maintenance tasks. Practical limitations, such as time and capacity, are considered through a case study involving a well-established building management organization. Results indicate that the Nearest Neighbor Algorithm generates 16–18 routes, the Saving Algorithm produces 18 routes, and the DRSO-driven EA also yields 18 routes. Significantly, the DRSO-driven EA consistently outperforms traditional methods, achieving a remarkable 17.7% reduction in route distance and an 18.8% reduction in journey time. In specific scenarios with 50 and 80 locations in Northeast and Central Bangkok, the DRSO-driven EA demonstrates practicality and efficacy. The algorithm's ability to address real-world challenges is underscored by these examples, showcasing its potential for broader implementation in real estate management. This study contributes significantly to the optimization of maintenance routes, presenting a clear roadmap for enhancing operational efficiency and customer satisfaction. Furthermore, it addresses the challenges posed by capitalism's growth constraints, offering insights for the adoption of sustainable economic practices in management, economics, and engineering.

Cosmic evolution early release science survey (CEERS): multiclassing galactic dwarf stars in the deep JWST/NIRCam

ABSTRACT Low-mass (sub)stellar objects represent the low end of the initial mass function, the transition to free-floating planets and a prominent interloper population in the search for high-redshift galaxies. To what accuracy can photometry only classify these? JWST/NIRCam has several advantages over Hubble Space Telescope (HST)/WFC3 near-infrared (NIR): more filters, a greater wavelength range, and greater spatial resolution. Here, we present a catalogue of (sub)stellar dwarfs identified in the cosmic evolution early release science survey (CEERS). We identify 518 stellar objects down to mF200W 28 using half-light radius, a full three magnitudes deeper than typical HST/WFC3 images. A k-means nearest neighbour (kNN) algorithm identifies and types these sources, using four HST/WFC3 and four NIRCam filters, trained on IRTF/spex spectra of nearby brown dwarfs. The kNN with four neighbours classifies well within two subtypes: e.g. M2±2, achieving ∼95 per cent precision and recall. In CEERS, we find 9 M8±2, 2 L6±2, 1 T4±2, and 15 T8±2. We compare the observed long wavelength NIRCam colours – not used in the kNN – to those expected for brown dwarf atmospheric models. The NIRCam F356W−F444W and F410M−F444W colours are redder by a magnitude for the type assigned by the kNN, hinting at a wider variety of atmospheres for these objects. We find a 300–350 pc scale height for M6±2 dwarfs plus a second structural component and a 150–200 pc scale height for T6±2 type dwarfs, consistent with literature values. A single M8±2 dwarf is spectroscopically confirmed at 4 kpc distance.

Ensemble learning-assisted prediction of prolonged hospital length of stay after spine correction surgery: a multi-center cohort study

PurposeThis research aimed to develop a machine learning model to predict the potential risk of prolonged length of stay in hospital before operation, which can be used to strengthen patient management.MethodsPatients who underwent posterior spinal deformity surgery (PSDS) from eleven medical institutions in China between 2015 and 2022 were included. Detailed preoperative patient data, including demographics, medical history, comorbidities, preoperative laboratory results, and surgery details, were collected from their electronic medical records. The cohort was randomly divided into a training dataset and a validation dataset with a ratio of 70:30. Based on Boruta algorithm, nine different machine learning algorithms and a stack ensemble model were trained after hyperparameters tuning visualization and evaluated on the area under the receiver operating characteristic curve (AUROC), precision-recall curve, calibration, and decision curve analysis. Visualization of Shapley Additive exPlanations method finally contributed to explaining model prediction.ResultsOf the 162 included patients, the K Nearest Neighbors algorithm performed the best in the validation group compared with other machine learning models (yielding an AUROC of 0.8191 and PRAUC of 0.6175). The top five contributing variables were the preoperative hemoglobin, height, body mass index, age, and preoperative white blood cells. A web-based calculator was further developed to improve the predictive model's clinical operability.ConclusionsOur study established and validated a clinical predictive model for prolonged postoperative hospitalization duration in patients who underwent PSDS, which offered valuable prognostic information for preoperative planning and postoperative care for clinicians.Trial registration ClinicalTrials.gov identifier NCT05867732, retrospectively registered May 22, 2023, https://classic.clinicaltrials.gov/ct2/show/NCT05867732.

Maximizing the Volume of Collocated Data from Two Coordinated Suborbital Platforms

Abstract Suborbital (e.g., airborne) campaigns that carry advanced remote sensing and in situ payloads provide detailed observations of atmospheric processes, but can be challenging to use when it is necessary to geographically collocate data from multiple platforms that make repeated observations of a given geographic location at different altitudes. This study reports on a data collocation algorithm that maximizes the volume of collocated data from two coordinated suborbital platforms and demonstrates its value using data from the NASA Aerosol Cloud Meteorology Interactions Over the western Atlantic Experiment (ACTIVATE) suborbital mission. A robust data collocation algorithm is critical for the success of the ACTIVATE mission goal to develop new and improved remote sensing algorithms, and quantify their performance. We demonstrate the value of these collocated data to quantify the performance of a recently developed vertically resolved lidar + polarimeter–derived aerosol particle number concentration (Na) product, resulting in a range-normalized mean absolute deviation (NMAD) of 9% compared to in situ measurements. We also show that this collocation algorithm increases the volume of collocated ACTIVATE data by 21% compared to using only nearest-neighbor finding algorithms alone. Additional to the benefits demonstrated within this study, the data files and routines produced by this algorithm have solved both the critical collocation and the collocation application steps for researchers who require collocated data for their own studies. This freely available and open-source collocation algorithm can be applied to future suborbital campaigns that, like ACTIVATE, use multiple platforms to conduct coordinated observations, e.g., a remote sensing aircraft together with in situ data collected from suborbital platforms. Significance Statement This study describes a data collocation (i.e., selection) process that aims to maximize the volume of data identified to be simultaneously collected in time and space from two coordinated measurement platforms. The functional utility of the resultant dataset is also demonstrated by extending the validation of aerosol particle number concentration derived from standard lidar and polarimeter data products from a suborbital mission that used two aircraft platforms.

De-noising classification method for financial time series based on ICEEMDAN and wavelet threshold, and its application

This paper proposes a classification method for financial time series that addresses the significant issue of noise. The proposed method combines improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and wavelet threshold de-noising. The method begins by employing ICEEMDAN to decompose the time series into modal components and residuals. Using the noise component verification approach introduced in this paper, these components are categorized into noisy and de-noised elements. The noisy components are then de-noised using the Wavelet Threshold technique, which separates the non-noise and noise elements. The final de-noised output is produced by merging the non-noise elements with the de-noised components, and the 1-NN (nearest neighbor) algorithm is applied for time series classification. Highlighting its practical value in finance, this paper introduces a two-step stock classification prediction method that combines time series classification with a BP (Backpropagation) neural network. The method first classifies stocks into portfolios with high internal similarity using time series classification. It then employs a BP neural network to predict the classification of stock price movements within these portfolios. Backtesting confirms that this approach can enhance the accuracy of predicting stock price fluctuations.