Synthetic Minority Over-sampling Technique Algorithm Research Articles

Boosting the classification success in imbalanced data of bee larva cells

Selecting the appropriate honey harvesting method is crucial for sustainable beekeeping and optimal honey production. The use of primitive harvesting methods can lead to the death of bees and a decrease in honey yield. This study aims to address the issue of detecting and classifying young larvae on honeycombs. However, the area where young larvae are found is limited compared to other areas. In this study, the dataset obtained from honeycombs was imbalanced, which has used the Synthetic Minority Oversampling TEchnique (SMOTE) algorithm to balance it. The SMOTE algorithm is a synthetic data generation method. The balanced dataset was then used for classification processes with k-Nearest Neighbors algorithm (k-NN), Decision Trees, and Support Vector Machines. The evaluation of the classification results included the F1-Score, G-Mean, and AUC metrics. The results showed that the classification of the dataset balanced with synthetic data was more successful.

[Establishment and verification of invasion syndrome prediction model in patients with diabetes complicated with Klebsiella pneumoniae liver abscess].

Objective: To analyze the correlative factors of invasion syndrome in patients with diabetes complicated with Klebsiella pneumoniae liver abscess, and to construct and verify the online nomographic prediction model. Methods: A case control study. The clinical data of 213 diabetic patients with Klebsiella pneumoniae liver abscess admitted to the Third Affiliated Hospital of Soochow University from January 1, 2015 to December 31, 2021 were retrospectively analyzed. The patients were divided into the training set (149 cases) and the test set (64 cases) by stratified random sampling method at a ratio of 7∶3. Synthetic minority over-sampling technique(SMOTE) was used to process the imbalanced data, then Lasso regression was used to screen out the optimal feature variables in the training set and multivariate logistic regression model was used to construct the prediction model of invasion syndrome in patients with diabetes complicated with Klebsiella pneumoniae liver abscess, and verify it in the training set and test set. Receiver operating characteristic (ROC) curve, calibration curve and decision curve analysis (DCA) were used to evaluate the prediction efficiency of the model, and the simple and online interactive dynamic web page column graph was constructed. Results: Among the 213 patients, 60 were males and 153 were females, aged of (61.4±12.0) years. A total of 25(11.74%) diabetic patients with Klebsiella pneumoniae liver abscess developed invasion syndrome, which were included in divided into invasive K.pneumoniae liver abscesses syndrome (IKPLAS) group, and the other 188 cases were in without invasive K.pneumoniae liver abscesses syndrome (NIKPLAS) group. SMOTE algorithm was used for oversampling processing, so that the ratio of positive and negative samples was 1∶1. In the oversampling training set, 5 main risk factors were screened based on Lasso regression, namely fasting blood glucose (λ=0.063), hemoglobin (λ=-0.042), blood urea nitrogen (λ=-0.050), abscess size (λ=-0.025) and sequential organ failure assessment (SOFA) score (λ=0.450), respectively. Multivariate logistic regression model showed that fasting blood glucose (OR=1.20, 95%CI: 0.98-1.48, P=0.006), hemoglobin (OR=0.90, 95%CI: 0.86-0.95, P<0.001), blood urea nitrogen (OR=1.22, 95%CI: 1.03-1.43, P=0.017), abscess diameter (OR=0.76, 95%CI: 0.61-0.94, P=0.010), SOFA score (OR=3.08, 95%CI: 2.18-4.36, P<0.001) were associated with invasion syndrome in patients with diabetes complicated with Klebsiella pneumoniae liver abscess. The area under the curve of ROC in the training set was 0.966 (95%CI: 0.943-0.989), the sensitivity was 90.5%, and the specificity was 91.3%. The area under the curve of the validation set ROC was 0.946 (95%CI: 0.902-0.991), with a sensitivity of 79.6% and a specificity of 88.9%. The calibration curves drawn in the training set and the test set fit well with the ideal curve. DCA showed that the neomorph prediction model had a good clinical net benefit when predicting the risk of IKPLAS in patients with diabetes complicated with Klebsiella pneumoniae liver abscess was 0.10-0.40. Conclusions: Fasting blood glucose, hemoglobin, urea nitrogen, abscess size and SOFA score are the related factors for invasion syndrome in patients with diabetes complicated with Klebsiella pneumoniae liver abscess. The constructed column graph can effectively predict the risk of invasion syndrome in patients with diabetes complicated with Klebsiae pneumoniae liver abscess.

Bank Customer Churn Prediction Using SMOTE: A Comparative Analysis

In today's market, customers have a plethora of options available to them when deciding where to invest their money. Consequently, customer churn and engagement have emerged as prominent concerns. With an increasing number of service providers targeting the same customer base, it is imperative for providers to understand evolving customer behavior and heightened expectations to retain their clientele. Numerous studies have addressed the issue of customer churn, with data mining frequently employed to predict bank customer attrition. While many researchers have proposed various approaches for predicting customer churn, some machine learning (ML) algorithms have struggled to deliver the required performance in identifying customer churn accurately most especially when the dataset is imbalance data. Therefore, this paper presents an application of Synthetic Minority Over Sampling Technique (SMOTE) on bank churn dataset. The SMOTE algorithm was employed to address the problem of data imbalance and Genetic Algorithm (GA) was applied to select most informative features from the original dataset. The selective features were evaluate using four (4) different classification algorithms: Random Forest (RF), K-Nearnest Neighbor (KNN), Artificial Neural Network (ANN) and Adaboost algorithms. The KNN model demonstrated superior performance compared to other models in terms of accuracy (96%), precision (96%), and F-measure (96%) respectively. Furthermore, we compared our results with existing models that utilized the same dataset, and our proposed strategy outperformed them.

Classification of recurrent major depressive disorder using a residual denoising autoencoder framework: Insights from large-scale multisite fMRI data

Background and objectiveRecurrent major depressive disorder (rMDD) has a high recurrence rate, and symptoms often worsen with each episode. Classifying rMDD using functional magnetic resonance imaging (fMRI) can enhance understanding of brain activity and aid diagnosis and treatment of this disorder. MethodsWe developed a Residual Denoising Autoencoder (Res-DAE) framework for the classification of rMDD. The functional connectivity (FC) was extracted from fMRI data as features. The framework addresses site heterogeneity by employing the Combat method to harmonize feature distribution differences. A feature selection method based on Fisher scores was used to reduce redundant information in the features. A data augmentation strategy using a Synthetic Minority Over-sampling Technique algorithm based on Extended Frobenius Norm measure was incorporated to increase the sample size. Furthermore, a residual module was integrated into the autoencoder network to preserve important features and improve the classification accuracy. ResultsWe tested our framework on a large-scale, multisite fMRI dataset, which includes 189 rMDD patients and 427 healthy controls. The Res-DAE achieved an average accuracy of 75.1 % (sensitivity = 69 %, specificity = 77.8 %) in cross-validation, thereby outperforming comparison methods. In a larger dataset that also includes first-episode depression (comprising 832 MDD patients and 779 healthy controls), the accuracy reached 70 %. ConclusionsWe proposed a deep learning framework that can effectively classify rMDD and 33 identify the altered FC associated with rMDD. Our study may reveal changes in brain function 34 associated with rMDD and provide assistance for the diagnosis and treatment of rMDD.

A Synthetic Minority Oversampling Technique Based on Gaussian Mixture Model Filtering for Imbalanced Data Classification.

Data imbalance is a common phenomenon in machine learning. In the imbalanced data classification, minority samples are far less than majority samples, which makes it difficult for minority to be effectively learned by classifiers. A synthetic minority oversampling technique (SMOTE) improves the sensitivity of classifiers to minority by synthesizing minority samples without repetition. However, the process of synthesizing new samples in the SMOTE algorithm may lead to problems such as "noisy samples" and "boundary samples." Based on the above description, we propose a synthetic minority oversampling technique based on Gaussian mixture model filtering (GMF-SMOTE). GMF-SMOTE uses the expected maximum algorithm based on the Gaussian mixture model to group the imbalanced data. Then, the expected maximum filtering algorithm is used to filter out the "noisy samples" and "boundary samples" in the subclasses after grouping. Finally, to synthesize majority and minority samples, we design two dynamic oversampling ratios. Experimental results show that the GMF-SMOTE performs better than the traditional oversampling algorithms on 20 UCI datasets. The population averages of sensitivity and specificity indexes of random forest (RF) on the UCI datasets synthesized by GMF-SMOTE are 97.49% and 97.02%, respectively. In addition, we also record the G-mean and MCC indexes of the RF, which are 97.32% and 94.80%, respectively, significantly better than the traditional oversampling algorithms. More importantly, the two statistical tests show that GMF-SMOTE is significantly better than the traditional oversampling algorithms.

Rapid and nondestructive THz inspection of unsound kernel of sunflower seed based on SMOTE algorithm

Inspection of unsound kernel of sunflower seed do an important role to the safe storage of sun-flower seeds and the quality of its processed products. A rapid and nondestructive method based on terahertz (THZ) time-domain spectroscopy was studied here to detect unsound kernel of sunflower seed. To explore the feasibility, the THz spectra of 446 sunflower seed samples were obtained. The synthetic minority over-sampling technique (SMOTE) was applied to balance the number of normal and unsound samples in the training set. To test effectiveness of SMOTE, the proportion of normal and unsound samples was set 4:1, 2:1 and 1.5:1 respectively in the experiment, then, new unsound samples was created by SMOTE algorithm till the number of unsound same to the normal, finally, the least squares support vector machine (LS-SVM) was used to construct the discrimination model. Comparing the performance of the model used SMOTE before and after, the sensitivity in the test set was improved from 0 % to 91.04 %, from 34.52 % to 92.86 %, and from 51.85 % to 92.59 % respectively. Five independent experiments with new sets of sample were repeated to verify that the model is robust, and the accuracy fluctuation is not more than ±5 %. THz time-domain spectrum, combined with SMOTE algorithm can effectively solve the problem of sample imbalance, thereby improving model performance. It is expected to provide feasible reference for accurate, nondestructive and rapid inspection of sunflower seeds and other shell agricultural products.

Cervical Cancer Prediction Using SMOTE Algorithm and Machine Learning Approaches

Cervical cancer is one of the most successful types of treatment when diagnosed early. In this study, it is aimed to find and classify the disease with data mining methods on the digitized data set obtained as a result of the pap-smear test. Two-stage architecture has been proposed for the diagnosis of cervical cancer. In the first stage of the study, missing data were extracted from the used dataset, and in the second stage, a new dataset was obtained by using the Synthetic Minority Oversampling Technique (SMOTE) algorithm to balance the target classes in the dataset. By applying the majority voting (MV) method to the dataset used in the study, the structure with 4 target variables was reduced to a single target variable. On two data sets, Artificial Neural Network (ANN), Support Vector Machines (SVM), Decision Trees (DT), Random Forest (RF), and K-Nearest Neighbors (KNN) algorithms from data mining methods were used for the diagnosis of cervical cancer. The results obtained from the original dataset and the dataset produced with Smote were compared. ANN is the best method evaluated according to classification success and F-score, and the major voted target variable in the balanced data group produced with the Smote algorithm gave the most successful result. The experimental results showed that the use of MV and SMOTE algorithms together increased the classification success from 93% to 99%.

Utilization of synthetic minority oversampling technique for improving potato yield prediction using remote sensing data and machine learning algorithms with small sample size of yield data

In recent years, the integration of machine learning (ML) algorithms and remote sensing data has been the commonly deployed practice for potato yield prediction in different scales. Since the quantity and quality of training data significantly affect ML algorithms' applicability, their effective use in some cases can be challenging and expensive. In this paper, we utilized the synthetic minority oversampling technique (SMOTE) algorithm to generate synthetic data for potato yield prediction. We conducted several experiments in two study sites called CS1 and CS2. The SMOTE algorithm was employed to produce synthetic data at five multiplication rates (5, 10, 20, 40, and 80). Six ML algorithms including random forest regression (RFR), support vector regression (SVR), K- nearest neighbor (KNN), extreme gradient boosting (XGB), deep neural network (DNN), and stacked auto-encoder of neural network (SAE) were used for potato yield prediction. To train the ML algorithms, multiple sets of synthetically generated data were combined with the original data. The similarity of synthetic data and original data was evaluated by two metrics (Kullback-Leibler divergence (KLD) and Jensen-Shannon divergence (JSD)), as well as PCA-based visualization. On the other hand, the root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) metrics were calculated to evaluate the performance of ML algorithms in potato yield prediction. Both quantitative and visual evaluations showed close similarity between the synthetic and original data. The average JSD (KLD) in CS1 and CS2 were 0.00028 (0.0031) and 0.161 (0.271), respectively. The ML algorithms showed noticeable differences when it comes to utilizing synthetic data. The RFR, XGB, DNN, and SAE algorithms positively responded to the addition of synthetic data, while SVR and KNN were the only ML algorithms that negatively responded to the addition of synthetic data. The DNN algorithm exhibited the highest positive response to the addition of synthetic data with an average RMSE change of −2.35 point percentage in CS1 and −24.54 point percentage in CS2. Although none of the ML algorithms and synthetic sample sizes provided the highest prediction performance in all the settings, which was plausible given the inherent differences among the selected ML algorithms, the RFR algorithm trained with the combination of original and quintupled synthetic data was the most appropriate choice for potato yield prediction.

A CT-based radiomics nomogram for distinguishing between malignant and benign Bosniak IIF masses: a two-centre study

To establish and assess a computed tomography (CT)-based radiomics nomogram for identifying malignant and benign Bosniak IIF masses. In total, 150 patients with Bosniak IIF masses were separated into a training set (n=106) and a test set (n=44) in a ratio of 7:3. A radiomics signature was calculated based on extracted features from the three phases of CT images. A clinical model was constructed based on clinical characteristics and CT features, and a nomogram incorporating the radiomics signature and independent clinical variables was established. The calibration ability, discrimination accuracy, and clinical value of the nomogram model were assessed. Twelve features derived from CT images were applied to establish the radiomics signature. The performance levels of three machine-learning models were improved by adding the synthetic minority oversampling technique algorithm. The optimised machine learning model was a combination of the minimum redundancy maximum relevance-least absolute shrinkage and selection operator feature screening method+logistic regression classifier+synthetic minority oversampling technique algorithm, which demonstrated excellent identification ability on the test set (area under the curve [AUC], 0.970; 95% confidence interval [CI], 0.940-1.000). The nomogram model displayed outstanding discrimination ability on the test set (AUC, 0.972; 95% CI, 0.942-1.000). The CT-based radiomics nomogram was useful for discriminating between malignant and benign Bosniak IIF masses, which improved the precision of preoperative diagnosis.

Glass transition of amorphous polymeric materials informed by machine learning

The glass transition temperature (Tg) is used to determine thermophysical properties of polymer materials and is often considered one of the most important descriptors. Methods for predicting various physical properties of materials based on machine learning algorithms and key molecular descriptors are efficient and accurate. However, it still needs improvements because an overly complex model is less practical and difficult to generalize. In addition, obtaining a large number of samples to achieve accurate predictions remains a challenge due to the complex and lengthy experimental process. In this work, based on Tg of 100 polymers, we use a feature selection algorithm combining FeatureWiz and the least absolute shrinkage and selection operator to quickly select molecular descriptors that are minimally redundant and maximally relevant to Tg. The processed dataset is interpolated from the original dataset using the nearest neighbor interpolation algorithm to solve the data deficiency problem. Finally, the synthetic minority oversampling technique algorithm is used to solve the data imbalance problem. The augmented dataset is used to construct the extreme gradient boosting prediction model to achieve good prediction accuracy. The experimental results demonstrate the robustness of the proposed model and the accuracy of its prediction results.

All-sky longwave radiation modelling based on infrared images and machine learning

Sky longwave radiation is an important input parameter for heat transfer and radiant energy balance calculations of the building envelope, and it has a significant effect on the accuracy of predictions of the energy consumption of buildings. The ability of the sky to emit longwave radiation needs to be accurately quantified under different weather conditions. In this study, a low-cost sky cloudiness observation platform based on an infrared imager was established, and an algorithm for calculating the cloudiness of the sky dome by stitching and processing infrared images was proposed. A machine learning model was developed by collecting local meteorological data; these data were then input into the synthetic minority oversampling technique algorithm. This model can estimate sky cloudiness based on conventional meteorological parameters. Nine different machine learning algorithms were tested, and a cloudiness prediction model based on the XGBoost algorithm was finally established, which had a prediction accuracy of 88.81%. Furthermore, an all-sky longwave radiation model was developed by introducing cloudiness parameters. The applicability of different longwave radiation models for clear and cloudy skies in subtropical climates was examined, and the coefficient values of the different models were modified. Finally, the formula applicable to subtropical climates was determined via a fitting method.

Abstract C013: Racial difference in risk factors for monoclonal gammopathy of undetermined significance – The premalignant stage of multiple myeloma

Abstract Background: Black race and obesity are known risk factors for type 2 diabetes mellitus (T2DM), multiple myeloma and its precursor condition – monoclonal gammopathy of undetermined significance (MGUS). Yet, little is known about what contributes to this racial difference, particularly in MGUS, and studies have inconsistently shown that T2DM may not be a risk factor for myeloma. We aimed to investigate the racial differences in risk factors for MGUS in patients with T2DM. Methods: Patients diagnosed with T2DM from 1999-2021 in the Veterans Health Administration were identified. We used a natural language processing-based algorithm to confirm MGUS diagnosis. Among them, patients of Black and White races with ≥2 BMI measurements were included. To preserve the prevalence of MGUS in the respective population, each Black patient with MGUS was matched to 27 Black patients without MGUS; and each White patient with MGUS was matched to 44 White patients without MGUS on BMI observation window. BMI observation window for each MGUS patient was defined as time from the 1st to the last BMI measurement before MGUS diagnosis. For the matched patients without MGUS, a similar length of observation window was selected along with ≥1 measurement beyond the window to ensure no loss to follow-up during the window. Each cohort was further split into the training (80%) and testing (20%) sets. Ad-hoc features for BMI in different statistics to summarize repeated measurements within the window were defined: average, variance (var), maximum (max), minimum (min), max BMI increase per month (max dBMI/m), max BMI decrease per month (min dBMI/m), and area under the dBMI curve. Ad-hoc features for other variables, HbA1c, creatinine, potassium, and glucose, were constructed in a similar fashion. To overcome the imbalanced MGUS cases (vs no MGUS), Synthetic Minority Over-Sampling Technique algorithm was used. To evaluate performance, we implemented 5-fold cross-validation using logistic regression and generated AUC score. Results: A total of 2,468 White and 3,981 Black T2DM MGUS patients were included. Logistic regression achieved an average AUC score of 0.76 (95% CI 0.72, 0.79) for Whites and 0.76 (95% CI 0.73, 0.78) for Blacks. In both races, the most significant risk factor for MGUS was max dBMI/m, followed by max BMI for Whites and var BMI for Blacks. Conversely, min glucose was the most significant factor inversely associated with the incidence of MGUS for both races. Conclusion: There is no significant difference in the most significant factors contributing to MGUS between the two races. In patients with T2DM, drastic change in body weight is the most significant risk factor for MGUS among others, while maintaining low glucose is the most protective factor for both races. Regardless of race, T2DM patients of high myeloma risk are encouraged to maintain a stable weight and low glucose level. Citation Format: Yao-Chi Yu, Mei Wang, Akhil Kumar, Kristin Vargo, Lawrence Liu, Theodore Thomas, Kristen Sanfilippo, Martin Schoen, Graham Colditz, Jr-Shin Li, Su-Hsin Chang. Racial difference in risk factors for monoclonal gammopathy of undetermined significance – The premalignant stage of multiple myeloma [abstract]. In: Proceedings of the 15th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2022 Sep 16-19; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2022;31(1 Suppl):Abstract nr C013.

Preliminary Construction of Prognostic Risk Early Warning Model for Renal Cell Carcinoma with Type 2 Diabetes Mellitus Based on SMOTE Algorithm

157 patients with renal cell carcinoma who underwent surgical treatment in our hospital from January 2019 to December 2020 were selected as the study subjects. Patients were divided into good prognosis group and poor prognosis group according to whether there were end-point events (death or distant metastasis or recurrence of tumor) during the follow-up period. The clinical data of the two groups of patients were collected, the independent risk factors for poor prognosis were screened by univariate and binary logistic regression analysis and the logistic regression model (P1) were established. At the same time, the data set was improved based on synthetic minority oversampling technique algorithm and the early warning model (P2) of the improved data set was constructed and the prediction efficiency of the model was compared and verified. There were 27 patients with end-point events during the follow-up period. The increased course of type 2 diabetes mellitus, high preoperative haemoglobin A1C, high body mass index and tumor–node–metastasis stage T3/T4 were independent risk factors for poor prognosis in patients with type 2 diabetes mellitus and renal cell carcinoma (p<0.05), while metformin was an independent protective factor for poor prognosis (p<0.05). Early warning model based on synthetic minority oversampling technique, oversampling algorithm P2=1/[1+e-(-13.084-0.438*X1+0.446*X2+0.096*X3-0.781*X4+1.155*X5)], determination coefficient and receiver operating characteristic of P2 model, the areas under the curve were significantly higher than those of the P1 model. The course of type 2 diabetes mellitus, preoperative, body mass index level, tumor–node–metastasis stage and whether to receive metformin hypoglycemic therapy in patients with type 2 diabetes mellitus combined with renal cell carcinoma are closely related to poor prognosis. Based on this, the individualized early warning model established by synthetic minority oversampling technique oversampling algorithm is beneficial to patients with high risk of poor prognosis early identification.

Fault diagnosis of power transformers using ANN and SMOTE algorithm

An artificial neural network (ANN) based methodology to diagnose transformer faults is proposed. The synthetic minority over-sampling technique (SMOTE) is used to solve the imbalance in the dataset. The SMOTE is improved by introducing a full cycle of creating synthetic samples from minority class samples for the goals that the over-sampled ratio can be automatically determined and the sample size of each category can be completely consistent. The contents of dissolved gases in transformer oils are treated as the original features. The optimal features combination for ANN is determined by comparing the performances of the ANN when different feature combinations are used. The performances of different activation functions used in the ANN are investigated to give the optimal one. The tested results show the high accuracy (97.92%) of the proposed methodology if the optimum feature combination and activation function are used.

Deep learning approach for cancer subtype classification using high-dimensional gene expression data

MotivationStudies have shown that classifying cancer subtypes can provide valuable information for a range of cancer research, from aetiology and tumour biology to prognosis and personalized treatment. Current methods usually adopt gene expression data to perform cancer subtype classification. However, cancer samples are scarce, and the high-dimensional features of their gene expression data are too sparse to allow most methods to achieve desirable classification results.ResultsIn this paper, we propose a deep learning approach by combining a convolutional neural network (CNN) and bidirectional gated recurrent unit (BiGRU): our approach, DCGN, aims to achieve nonlinear dimensionality reduction and learn features to eliminate irrelevant factors in gene expression data. Specifically, DCGN first uses the synthetic minority oversampling technique algorithm to equalize data. The CNN can handle high-dimensional data without stress and extract important local features, and the BiGRU can analyse deep features and retain their important information; the DCGN captures key features by combining both neural networks to overcome the challenges of small sample sizes and sparse, high-dimensional features. In the experiments, we compared the DCGN to seven other cancer subtype classification methods using breast and bladder cancer gene expression datasets. The experimental results show that the DCGN performs better than the other seven methods and can provide more satisfactory classification results.

Development and validation of prediction models for hypertension risks: A cross-sectional study based on 4,287,407 participants.

ObjectiveTo develop an optimal screening model to identify the individuals with a high risk of hypertension in China by comparing tree-based machine learning models, such as classification and regression tree, random forest, adaboost with a decision tree, extreme gradient boosting decision tree, and other machine learning models like an artificial neural network, naive Bayes, and traditional logistic regression models.MethodsA total of 4,287,407 adults participating in the national physical examination were included in the study. Features were selected using the least absolute shrinkage and selection operator regression. The Borderline synthetic minority over-sampling technique was used for data balance. Non-laboratory and semi-laboratory analyses were carried out in combination with the selected features. The tree-based machine learning models, other machine learning models, and traditional logistic regression models were constructed to identify individuals with hypertension, respectively. Top features selected using the best algorithm and the corresponding variable importance score were visualized.ResultsA total of 24 variables were finally included for analyses after the least absolute shrinkage and selection operator regression model. The sample size of hypertensive patients in the training set was expanded from 689,025 to 2,312,160 using the borderline synthetic minority over-sampling technique algorithm. The extreme gradient boosting decision tree algorithm showed the best results (area under the receiver operating characteristic curve of non-laboratory: 0.893 and area under the receiver operating characteristic curve of semi-laboratory: 0.894). This study found that age, systolic blood pressure, waist circumference, diastolic blood pressure, albumin, drinking frequency, electrocardiogram, ethnicity (uyghur, hui, and other), body mass index, sex (female), exercise frequency, diabetes mellitus, and total bilirubin are important factors reflecting hypertension. Besides, some algorithms included in the semi-laboratory analyses showed less improvement in the predictive performance compared to the non-laboratory analyses.ConclusionUsing multiple methods, a more significant prediction model can be built, which discovers risk factors and provides new insights into the prediction and prevention of hypertension.

Discrimination of Chemical Oxygen Demand Pollution in Surface Water Based on Visible Near-Infrared Spectroscopy

Chemical oxygen demand (COD) is one of the indicators used to monitor the level of pollution in surface water. To recycle agricultural water resources, it is crucial to monitor, in a timely manner, whether COD in surface water exceeds the agricultural water control standard. A diagnostic model of surface water pollution was developed using visible near-infrared spectroscopy (Vis-NIR) combined with partial least squares discriminant analysis (PLS–DA). A total of 127 surface water samples were collected from Guangzhou, Guangdong, China. The COD content was measured using the potassium dichromate method. The spectra of the surface water samples were recorded using a Vis-NIR spectrometer, and the spectral data were pre-processed using four different methods. To improve the accuracy and simplicity of the model, the synthetic minority oversampling technique (SMOTE) and the competitive adaptive reweighted sampling (CARS) algorithm were used to enhance model performance. The best PLS–DA model achieved an accuracy of 88%, and the SMOTE–PLS–DA model had an accuracy of 94%. The SMOTE algorithm could improve the accuracy of the model despite the sampling imbalance. The CARS–SMOTE–PLS–DA model achieved 97% accuracy, and the CARS band selection technique improved the simplicity and accuracy of the discrimination model. The CARS–SMOTE–PLS–DA model improved the discrimination accuracy by 9% over that of the PLS–DA model. This method can not only save human and material resources but is also a new way for real-time online discrimination of COD in surface water.

The Application of Machine Learning in Predicting Mortality Risk in Patients With Severe Femoral Neck Fractures: Prediction Model Development Study

Background Femoral neck fracture (FNF) accounts for approximately 3.58% of all fractures in the entire body, exhibiting an increasing trend each year. According to a survey, in 1990, the total number of hip fractures in men and women worldwide was approximately 338,000 and 917,000, respectively. In China, FNFs account for 48.22% of hip fractures. Currently, many studies have been conducted on postdischarge mortality and mortality risk in patients with FNF. However, there have been no definitive studies on in-hospital mortality or its influencing factors in patients with severe FNF admitted to the intensive care unit. Objective In this paper, 3 machine learning methods were used to construct a nosocomial death prediction model for patients admitted to intensive care units to assist clinicians in early clinical decision-making. Methods A retrospective analysis was conducted using information of a patient with FNF from the Medical Information Mart for Intensive Care III. After balancing the data set using the Synthetic Minority Oversampling Technique algorithm, patients were randomly separated into a 70% training set and a 30% testing set for the development and validation, respectively, of the prediction model. Random forest, extreme gradient boosting, and backpropagation neural network prediction models were constructed with nosocomial death as the outcome. Model performance was assessed using the area under the receiver operating characteristic curve, accuracy, precision, sensitivity, and specificity. The predictive value of the models was verified in comparison to the traditional logistic model. Results A total of 366 patients with FNFs were selected, including 48 cases (13.1%) of in-hospital death. Data from 636 patients were obtained by balancing the data set with the in-hospital death group to survival group as 1:1. The 3 machine learning models exhibited high predictive accuracy, and the area under the receiver operating characteristic curve of the random forest, extreme gradient boosting, and backpropagation neural network were 0.98, 0.97, and 0.95, respectively, all with higher predictive performance than the traditional logistic regression model. Ranking the importance of the feature variables, the top 10 feature variables that were meaningful for predicting the risk of in-hospital death of patients were the Simplified Acute Physiology Score II, lactate, creatinine, gender, vitamin D, calcium, creatine kinase, creatine kinase isoenzyme, white blood cell, and age. Conclusions Death risk assessment models constructed using machine learning have positive significance for predicting the in-hospital mortality of patients with severe disease and provide a valid basis for reducing in-hospital mortality and improving patient prognosis.

Intrusion detection system combined enhanced random forest with SMOTE algorithm

Network security is subject to malicious attacks from multiple sources, and intrusion detection systems play a key role in maintaining network security. During the training of intrusion detection models, the detection results generally have relatively large false detection rates due to the shortage of training data caused by data imbalance. To address the existing sample imbalance problem, this paper proposes a network intrusion detection algorithm based on the enhanced random forest and synthetic minority oversampling technique (SMOTE) algorithm. First, the method used a hybrid algorithm combining the K-means clustering algorithm with the SMOTE sampling algorithm to increase the number of minor samples and thus achieved a balanced dataset, by which the sample features of minor samples could be learned more effectively. Second, preliminary prediction results were obtained by using enhanced random forest, and then the similarity matrix of network attacks was used to correct the prediction results of voting processing by analyzing the type of network attacks. In this paper, the performance was tested using the NSL-KDD dataset with a classification accuracy of 99.72% on the training set and 78.47% on the test set. Compared with other related papers, our method has some improvement in the classification accuracy of detection.

A Machine Learning Framework to Predict the Tensile Stress of Natural Rubber: Based on Molecular Dynamics Simulation Data.

Natural rubber (NR), with its excellent mechanical properties, has been attracting considerable scientific and technological attention. Through molecular dynamics (MD) simulations, the effects of key structural factors on tensile stress at the molecular level can be examined. However, this high-precision method is computationally inefficient and time-consuming, which limits its application. The combination of machine learning and MD is one of the most promising directions to speed up simulations and ensure the accuracy of results. In this work, a surrogate machine learning method trained with MD data is developed to predict not only the tensile stress of NR but also other mechanical behaviors. We propose a novel idea based on feature processing by combining our previous experience in performing predictions of small samples. The proposed ML method consists of (i) an extreme gradient boosting (XGB) model to predict the tensile stress of NR, and (ii) a data augmentation algorithm based on nearest-neighbor interpolation (NNI) and the synthetic minority oversampling technique (SMOTE) to maximize the use of limited training data. Among the data enhancement algorithms that we design, the NNI algorithm finally achieves the effect of approaching the original data sample distribution by interpolating at the neighborhood of the original sample, and the SMOTE algorithm is used to solve the problem of sample imbalance by interpolating at the clustering boundaries of minority samples. The augmented samples are used to establish the XGB prediction model. Finally, the robustness of the proposed models and their predictive ability are guaranteed by high performance values, which indicate that the obtained regression models have good internal and external predictive capacities.