Synthetic Minority Over-sampling Technique Technique Research Articles

Binding Activity Classification of Anti-SARS-CoV-2 Molecules using Deep Learning Across Multiple Assays

The coronavirus disease-2019 (COVID-19) pandemic, caused by severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2), has urgently necessitated effective therapeutic solutions, with a focus on rapidly identifying and classifying potential small-molecule drugs. Given traditional methods’ labor-intensive and time-consuming nature, deep learning has emerged as an essential tool for efficiently processing and extracting insights from complex biological data. To utilize deep learning techniques, particularly deep neural networks (DNN) enhanced with the synthetic minority oversampling technique (SMOTE), to enhance the classification of binding activities in anti-SARS-CoV-2 molecules across various bioassays. We used 11 bioassay datasets covering various SARS-CoV-2 interactions and inhibitory mechanisms. These assays ranged from spike-ACE2 protein-protein interaction to ACE2 enzymatic activity and 3CL enzymatic activity. To address the prevalent class imbalance in these datasets, the SMOTE technique was employed to generate new samples for the minority class. In our model-building approach, we divided the dataset into 80% training and 20% test sets, reserving 10% of the training set for validation. Our approach involved employing a DNN that integrates ReLU and sigmoid activation functions, incorporates batch normalization, and uses Adam optimization. The hyperparameters and architecture of the DNN were optimized through various tests on layers, minibatch sizes, epoch sizes, and learning rates. A 40% dropout rate was incorporated to mitigate overfitting. For model evaluation, we computed performance metrics, such as balanced accuracy (BACC), precision, recall, F1 score, Matthews’ correlation coefficient (MCC), and area under the curve (AUC). The performance of the DNN across 11 bioassay test sets revealed varying outcomes, significantly influenced by the ratios of active-to-inactive compounds. Assays, such as AlphaLISA and CoV-PPE, demonstrated robust performance across various metrics, including BACC, precision, recall, and AUC, when configured with more balanced ratios (1:3 and 1:1, respectively). This suggests the effective identification of active compounds in both cases. In contrast, assays with higher imbalance ratios, such as 3CL (1:38) and cytopathic effect (1:15), demonstrated higher recall but lower precision, highlighting challenges in accurately identifying active compounds among numerous inactive compounds. However, even in these challenging settings, the model achieved favorable BACC and recall scores. Overall, the DNN model generally performed well, as indicated by the BACC, MCC, and AUC values, especially when considering the degree of dataset imbalance in each assay. This study demonstrates the significant impact of deep learning, particularly DNN models enhanced with SMOTE, in improving the identification of active compounds in bioassay datasets for COVID-19 drug discovery, outperforming traditional machine learning models. Furthermore, this study highlights the efficacy of advanced computational techniques in addressing high-throughput screening data imbalances.

Improving the performance of machine learning model predicting phase and crystal structure of high entropy alloys by the synthetic minority oversampling technique

It is critical to select the suitable integration of elements that has an influence on phase formation when selecting the proper High Entropy Alloys (HEAs) with desired qualities. Machine Learning (ML) models were used to predict the phase and crystal structure of HEAs in this study. Trained on extensive experimental datasets, the set of input characteristics includes Atomic size difference, Mixing Enthalpy, Mixing Entropy, Valence Electron concentration, Electronegativity difference, Ω and Average Melting Temperature. The phase is divided into three categories: Solid solution (S), Dual phase (D), and Intermetallic (Im). Crystal structure classifications include BCC, FCC, and FCC+BCC. For phase and crystal structure prediction, five ML algorithms were used: Decision Tree (DT), K Nearest Neighbour (KNN), Random Forest (RF), Gradient Boosting (GB) and eXtreme Gradient Boosting (XG Boosting), The classification of the HEAs phase has been predicted with great accuracy using an XG Boosting (80 %) and Random Forest exhibits the highest accuracy (94 %) for crystal structure prediction when model is trained on an experimental dataset. Even though the ML model predicts the classes with a better degree of accuracy (>80 %), the F1 value of each class varies more dramatically. This issue has arisen as a result of an imbalance in the dataset's class distribution. To boost performance, this work uses an ML model in conjunction with the Synthetic Minority Oversampling Technique (SMOTE). According to this approach, XG boosting identified the crystal structure with 93 % accuracy and predicted the phase of HEAs with 84 % accuracy. Additionally, it has been noted that each class's F1 score improved and the difference between them was reduced after using the SMOTE technique. It is evident that the SMOTE approach has improved the ML model's ability to differentiate between the various classes.

Predictive modeling for postoperative delirium in elderly patients with abdominal malignancies using synthetic minority oversampling technique.

Postoperative delirium, particularly prevalent in elderly patients after abdominal cancer surgery, presents significant challenges in clinical management. To develop a synthetic minority oversampling technique (SMOTE)-based model for predicting postoperative delirium in elderly abdominal cancer patients. In this retrospective cohort study, we analyzed data from 611 elderly patients who underwent abdominal malignant tumor surgery at our hospital between September 2020 and October 2022. The incidence of postoperative delirium was recorded for 7 d post-surgery. Patients were divided into delirium and non-delirium groups based on the occurrence of postoperative delirium or not. A multivariate logistic regression model was used to identify risk factors and develop a predictive model for postoperative delirium. The SMOTE technique was applied to enhance the model by oversampling the delirium cases. The model's predictive accuracy was then validated. In our study involving 611 elderly patients with abdominal malignant tumors, multivariate logistic regression analysis identified significant risk factors for postoperative delirium. These included the Charlson comorbidity index, American Society of Anesthesiologists classification, history of cerebrovascular disease, surgical duration, perioperative blood transfusion, and postoperative pain score. The incidence rate of postoperative delirium in our study was 22.91%. The original predictive model (P1) exhibited an area under the receiver operating characteristic curve of 0.862. In comparison, the SMOTE-based logistic early warning model (P2), which utilized the SMOTE oversampling algorithm, showed a slightly lower but comparable area under the curve of 0.856, suggesting no significant difference in performance between the two predictive approaches. This study confirms that the SMOTE-enhanced predictive model for postoperative delirium in elderly abdominal tumor patients shows performance equivalent to that of traditional methods, effectively addressing data imbalance.

Regional 3D geological modeling along metro lines based on stacking ensemble model

This paper presents a regional 3D geological modeling method based on the stacking ensemble technique to overcome the challenges of sparse borehole data in large-scale linear underground projects. The proposed method transforms the 3D geological modeling problem into a stratigraphic property classification problem within a subsurface space grid cell framework. Borehole data is pre-processed and trained using stacking method with five different machine learning algorithms. The resulting modelled regional cells are then classified, forming a regional 3D grid geological model. A case study for an area of 324 km2 along Xuzhou metro lines is presented to demonstrate the effectiveness of the proposed model. The study shows an overall prediction accuracy of 85.4%. However, the accuracy for key stratigraphy layers influencing the construction risk, such as karst carve strata, is only 4.3% due to the limited borehole data. To address this issue, an oversampling technique based on the synthetic minority oversampling technique (SMOTE) algorithm is proposed. This technique effectively increases the number of sparse stratigraphic samples and significantly improves the prediction accuracy for karst caves to 65.4%. Additionally, this study analyzes the impact of sampling distance on model accuracy. It is found that a lower sampling interval results in higher prediction accuracy, but also increases computational resources and time costs. Therefore, in this study, an optimal sampling distance of 1 m is chosen to balance prediction accuracy and computation cost. Furthermore, the number of geological strata is found to have a negative effect on prediction accuracy. To mitigate this, it is recommended to merge less significant stratigraphy layers, reducing computation time. For key strata layers, such as karst caves, which have a significant impact on construction risk, further on-site sampling or oversampling using the SMOTE technique is recommended.

An Oversampling Technique for Handling Imbalanced Data in Patients with Metabolic Syndrome and Periodontitis

Objectives: Periodontitis has been suggested to be associated with several systemic diseases and conditions including obesity, metabolic syndrome, diabetes, chronic renal disease, respiratory disorders, and cardiovascular diseases. Metabolic syndrome (MetS) is a collection of impairment and is a risk factor for type 2 diabetes and cardiovascular disease. Our study is aimed to handle MetS unbalanced data using the synthetic minority over-sampling technique (SMOTE) to increase accuracy and reliability.
 Materials and Methods: Six metabolic syndrome patients and 26 systemically healthy subjects with periodontitis were recruited in this study. Clinical parameters (Plaque index (PI), gingival index (GI), probing pocket depth (PPD), clinical attachment loss (CAL), and bleeding on probing (BOP)) were obtained, smoking status and body-mass index (BMI), systemic diseases, fasting glucose levels, hemoglobin A1c (HbA1c) levels and serum advanced glycation end-products (AGE) levels were recorded by one examiner. First, the data was pre-processed by removing missing values, outliers and normalizing the data. Then, SMOTE technique was used to oversample the minority class. SMOTE works by creating synthetic data points that are similar to the existing minority class instances. The experimental dataset included numerous machine learning algorithms and assessed accuracy using both pre- and post-oversampling methods.
 Results: Our findings suggest that by increasing the sample size of a study, researchers can gain more accurate and reliable results. This is especially important when studying a population with a lower sample size, as the results may be skewed.
 Conclusion: SMOTE may result in over fitting on numerous copies of minority class samples.

Implementation of Naïve Bayes Algorithm in Sentiment Analysis of Twitter Social Media Users Regarding Their Interest to Pay the Tax

Since 2008, tax revenue has failed to reach the target set in the State Budget each year. Until 2021, tax revenue managed to reach the target that had been targeted in the 2021 state budget. In the midst of improving tax revenue, towards the end of February 2023, a case involving the son of a Directorate General of Taxes (DGT) that made the father called by the Corruption Eradication Commission (CEC) to be asked for an explanation of his assets. After the case, there were many calls in the community to stop paying taxes, which was assessed by Tauhid Ahmad as Executive Director of Indef as a form of decreased trust in tax collecting institutions. This can affect the amount of revenue from taxes because trust in the government is one of the factors that tend to affect public compliance in paying taxes. Which can affect the amount of revenue from taxes because trust in the government is one of the factors that tend to affect public compliance in paying taxes. One of the crowded calls is the pros and cons of the tax boycott movement on Twitter. With the pros and cWith the pros and cons of the movement that can affect tax revenues on Twitter social media, an assessment based on sentiment analysis is needed which is divided into positive, neutral, or negative categories. Sentiment analysis in this research is carried out using three variations of Naïve Bayes assisted by the TF-IDF word weighting model, namely Gaussian Naïve Bayes, Multinomial Naïve Bayes, and Bernoulli Naïve Bayes. Then Confussion Matrix is used to evaluate the model by obtaining the accuracy, precission, recall, and f1-score values and the use of Synthetic Minority Oversampling Technique (SMOTE) to handle unbalanced data. The results of this study on unbalanced data, the implementation of Bernoulli Naïve Bayes using the SMOTE technique on a dataset comparison of 80:20 resulted in better performance than the variations of Gaussian and Multinomial Naïve Bayes with accuracy results of 91.03%, precision, 71.11%, recall 71.43%, and f1-score of 71.18%.

An efficient ensemble based machine learning approach for predicting Chronic Kidney Disease.

Chronic kidney disease (CKD) is a long-term risk to one's health that can result in kidney failure. CKD is one of today's most serious diseases, and early detection can aid in proper treatment. Machine learning techniques have proven to be reliable in the early medical diagnosis. The paper aims to perform CKD prediction using machine learning classification approaches. The dataset used for the present study for detecting CKD was obtained from the machine learning repository at the University of California, Irvine (UCI). In this study, twelve machine learning-based classification algorithms with full features were used. Since the CKD dataset had a class imbalance issue, the Synthetic Minority Over-Sampling technique (SMOTE) was used to alleviate the problem of class imbalance and review the performance based on machine learning classification models using the K fold cross-validation technique. The proposed work compares the results of twelve classifiers with and without the SMOTE technique, and then the top three classifiers with the highest accuracy, Support Vector Machine, Random Forest, and Adaptive Boosting classification algorithms were selected to use the ensemble technique to improve performance. The accuracy achieved using a stacking classifier as an ensemble technique with cross-validation is 99.5%. The study provides an ensemble learning approach in which the top three best-performing classifiers in terms of cross-validation results are stacked in an ensemble model after balancing the dataset using SMOTE. This proposed technique could be applied to other diseases in the future, making disease detection less intrusive and cost-effective.

Comparative analysis of resampling algorithms in the prediction of stroke diseases

Stroke disease is a serious cause of death globally. Early predictions of the disease will save a lot of lives but most of the clinical datasets are imbalanced in nature including the stroke dataset, making the predictive algorithms biased towards the majority class. The objective of this research is to compare different data resampling algorithms on the stroke dataset to improve the prediction performances of the machine learning models. This paper considered five (5) resampling algorithms namely; Random over Sampling (ROS), Synthetic Minority oversampling Technique (SMOTE), Adaptive Synthetic (ADASYN), hybrid techniques like SMOTE with Edited Nearest Neighbor (SMOTE-ENN), and SMOTE with Tomek Links (SMOTE-TOMEK) and trained on six (6) machine learning classifiers namely; Logistic Regression (LR), Decision Tree (DT), K-nearest Neighbor (KNN), Support Vector Machines (SVM), Random Forest (RF), and XGBoost (XGB). The hybrid technique SMOTE-ENN influences the machine learning classifiers the best followed by the SMOTE technique while the combination of SMOTE and XGB perform better with an accuracy of 97.99% and G-mean score of 0.99, and auc_roc score of 0.99. Resampling algorithms balance the dataset and enhanced the predictive power of machine learning algorithms. Therefore, we recommend resampling stroke dataset in predicting stroke disease than modeling on the imbalanced dataset.

Discovery of potential biomarkers for lung cancer classification based on human proteome microarrays using Stochastic Gradient Boosting approach.

Early identification of lung cancer (LC) will considerably facilitate the intervention and prevention of LC. The human proteome micro-arrays approach can be used as a "liquid biopsy" to diagnose LC to complement conventional diagnosis, which needs advanced bioinformatics methods such as feature selection (FS) and refined machine learning models. A two-stage FS methodology by infusing Pearson's Correlation (PC) with a univariate filter (SBF) or recursive feature elimination (RFE) was used to reduce the redundancy of the original dataset. The Stochastic Gradient Boosting (SGB), Random Forest (RF), and Support Vector Machine (SVM) techniques were applied to build ensemble classifiers based on four subsets. The synthetic minority oversampling technique (SMOTE) was used in the preprocessing of imbalanced data. FS approach with SBF and RFE extracted 25 and 55 features, respectively, with 14 overlapped ones. All three ensemble models demonstrate superior accuracy (ranging from 0.867 to 0.967) and sensitivity (0.917to 1.00) in the test datasets with SGB of SBF subset outperforming others. The SMOTE technique has improved the model performance in the training process. Three of the top selected candidate biomarkers (LGR4, CDC34, and GHRHR) were highly suggested to play a role in lung tumorigenesis. A novel hybrid FS method with classical ensemble machine learning algorithms was first used in the classification of protein microarray data. The parsimony model constructed by the SGB algorithm with the appropriate FS and SMOTE approach performs well in the classification task with higher sensitivity and specificity. Standardization and innovation of bioinformatics approach for protein microarray analysis need further exploration and validation.

Performance Analysis of Machine Learning Methods with Class Imbalance Problem in Android Malware Detection

Due to the exponential rise of mobile technology, a slew of new mobile security concerns has surfaced recently. To address the hazards connected with malware, many approaches have been developed. Signature-based detection is the most widely used approach for detecting Android malware. This approach has the disadvantage of being unable to identify unknown malware. As a result of this issue, machine learning (ML) for identifying and categorising malware apps was created. Conventional ML methods are concerned with increasing classification accuracy. However, the standard classification method performs poorly in recognising malware applications due to the unbalanced real-world datasets. In this study, an empirical analysis of the detection performance of ML methods in the presence of class imbalance is conducted. Specifically, eleven (11) ML methods with diverse computational complexities were investigated. Also, a synthetic minority oversampling technique (SMOTE) and random undersampling (RUS) are deployed to address the class imbalance in the Android malware datasets. The experimented ML methods are tested using the Malgenome and Drebin Android malware datasets that contain features gathered from both static and dynamic malware approaches. According to the experimental findings, the performance of each experimented ML method varies across the datasets. Moreover, the presence of class imbalance deteriorated the performance of the ML methods as their performances were amplified with the deployment of data sampling methods (SMOTE and RUS) used to alleviate the class imbalance problem. Besides, ML models with SMOTE technique are superior to other experimented methods. It is therefore recommended to address the inherent class imbalance problem in Android Malware detection.

Building vulnerability assessment in seismic areas using ensemble learning: A Nepal case study

The vulnerability assessment is a crucial method for earthquake-prone areas to manage seismic risk. This paper develops an Extreme gradient boosting (XGBoost)-based model to assess the seismic vulnerability of buildings in Kavrepalanchok, Nepal. The vulnerability assessment model is trained based on the building survey data of the 2015 Gorkha earthquake. To rectify the impacts of class imbalance in raw data, the Synthetic Minority Oversampling Technique (SMOTE) technique is adopted in the data preprocessing stage. Then, XGBoost models for the 3 selected target variables are developed respectively. Finally, the performances of the developed models are evaluated, and the significant features that influence the vulnerability are investigated using the SHapley Additive exPlanations (SHAP) technique. Important findings are: (1) The developed SMOTE-XGBoost model for the overall building damage achieves the highest macro f1 score of 0.729 compared with other machine learning models. (2) The SMOTE technique is favorable when the label requiring more attention is the minor group (e.g., severe damage). (3) The foundation type, the roof type, and the building age are identified as the most significant features for the overall building damage grade. The developed learning model can be used as a decision tool to automate building vulnerability assessment in seismic areas using pre-earthquake data, providing decision support for disaster risk reduction at an early stage.

Assessment of Factors Influencing the Survival of Breast Cancer Patients using a Machine Learning Approach

Breast cancer is one of the deadliest diseases, claiming approximately 627,000 lives worldwide in 2018–2019. Therefore, early detection of breast cancer through automation in the prediction of the disease will help the medical industry to cure this disease at an early stage and thereby reduce the risk of death drastically. In the present study, the Breast Cancer Wisconsin (Diagnostic) Data Set has been taken from the University of California Irvine (UCI) Machine Learning Repository. The dataset (n=699) contained a total of 30 predictor parameters and one dependent parameter. The dependent variable referred to the type of cancer tissue, i.e., benign or malignant. To predict the type of cancer tissue present in the patient, prediction models were built using 1) Logistic Regression (LR), 2) Decision Tree Classifier (DTC), 3) Random Forest Classifier (RFC), 4) K Nearest Neighbor (KNN), 5) Support Vector Machine (SVM), and 6) Ada Boost Classifier (ABC). To improve the accuracy of the model, a correlation matrix was used and the top 8 features were selected. To improve the accuracy even further, the Synthetic Minority Oversampling Technique (SMOTE) was used to eliminate the problem of class imbalance, and then accuracy was compared before and after SMOTE. The Precision, Recall, and F1 scores are the performance metrics that have been taken into consideration for selecting the best model for the analysis. The results of the study reveal that the KNN algorithm gives the highest accuracy of 95.321% after the SMOTE technique is applied to each of the six algorithms. It has been revealed that while SMOTE aids in the accuracy of some algorithms, it affects the performance of others. This research may be turned into realistic tools that can be utilized in the medical field to more accurately predict the stage of disease for better treatment management.

Penerapan Algoritma C4.5 Pada Imbalanced Dataset Untuk Memprediksi Kegagalan Angsuran Properti

In this research, the data collection carried out by studying the patterns of consumers who fail to pay, which aimed to build a model so that it could be used in predicting customers who have the potential to fail to pay. The research used the Cross-Industry Standard Process for Data Mining (CRISP-DM) method with details of the business understanding process, data understanding, data preparation, modeling, evaluation and deployment / interpretation. The dataset in this research was taken from sales, cancellation and consumer data from January 2016 to December 2019. Because the dataset in this research was an imbalanced dataset, the researchers tried to use Synthetic Minority Oversampling Technique (SMOTE) in handling the imbalanced dataset. The research conducted a comparison of the value of accuracy, precision, recall, f measure and Area Under the ROC Curve (AUC) between the original dataset and the dataset for the addition of the SMOTE technique to several algorithms including C4.5, K-NN and Naïve Bayes. The attributes used in this research were source of funds, purpose of purchase, age, selling price, occupation, total installments, percentage of total installments, monthly installments, percentage of late installments and status. From the comparison, it was found that the C4.5 algorithm with the SMOTE 480% dataset had the highest accuracy value of 97.62%, precision of 0.976, recall of 0.976, f measure of 0.976 and AUC of 0.986 which meant Excellent Classification. From the research conducted, it was expected that the model formed on the imbalanced dataset with the C4.5 and SMOTE algorithms could be used to predict consumer installment failures.

Detecting Electronic Banking Fraud on Highly Imbalanced Data using Hidden Markov Models

Recent researches have revealed the capability of Machine Learning (ML) techniques to effectively detect fraud in electronic banking transactions since they have the potential to detect new and unknown intrusions. A major challenge in the application of ML to fraud detection is the presence of highly imbalanced data sets. In many available datasets, majority of transactions are genuine with an extremely small percentage of fraudulent ones. Designing an accurate and efficient fraud detection system that is low on false positives but detects fraudulent activity effectively is a significant challenge for researchers. In this paper, a framework based on Hidden Markov Models (HMM), modified Density Based Spatial Clustering of Applications with Noise (DBSCAN) and Synthetic Minority Oversampling Technique Techniques (SMOTE) is proposed to effectively detect fraud in a highly imbalanced electronic banking dataset. The various transaction types, transaction amounts and the frequency of transactions are taken into consideration by the proposed model to enable effective detection. With different number of hidden states for the proposed HMMs, simulations are performed for four (4) different approaches and their performances compared using precision, recall rate and F1-Score as the evaluation metrics. The study revealed that, our proposed approach is able to detect fraudulent transactions more effectively with reasonably low number of false positives.

Exploration of Obesity Status of Indonesia Basic Health Research 2013 With Synthetic Minority Over-Sampling Techniques

The accuracy of the data class is very important in classification with a machine learning approach. The more accurate the existing data sets and classes, the better the output generated by machine learning. In fact, classification can experience imbalance class data in which each class does not have the same portion of the data set it has. The existence of data imbalance will affect the classification accuracy. One of the easiest ways to correct imbalanced data classes is to balance it. This study aims to explore the problem of data class imbalance in the medium case dataset and to address the imbalance of data classes as well. The Synthetic Minority Over-Sampling Technique (SMOTE) method is used to overcome the problem of class imbalance in obesity status in Indonesia 2013 Basic Health Research (RISKESDAS). The results show that the number of obese class (13.9%) and non-obese class (84.6%). This means that there is an imbalance in the data class with moderate criteria. Moreover, SMOTE with over-sampling 600% can improve the level of minor classes (obesity). As consequence, the classes of obesity status balanced. Therefore, SMOTE technique was better compared to without SMOTE in exploring the obesity status of Indonesia RISKESDAS 2013.

Machine Learning Models for Diagnostic Classification of Hepatitis C Tests

Introduction: Hepatitis C is a chronic infection caused by hepatitis c virus - a blood borne virus. Therefore, the infection occurs through exposure to small quantities of blood. It has been estimated by World Health Organization (WHO) to have affected 71 million people worldwide. This infection costs individual, groups and government a lot because no vaccine has been gotten yet for the treatment. This disease is likely to continue to affect more people because it’s long asymptotic phase which makes its early detection not feasible.Material and Methods: In this study, we have presented machine learning models to automatically classify the diagnosis test of hepatitis and also ranked the test features in order to know how they contribute to the classification which help in decision making process by the health care industry. The synthetic minority oversampling technique (SMOTE) was used to solve the problem of imbalance dataset.Results: The models were evaluated based on metrics such as Matthews correlation coefficient, F-measure, Precision-Recall curve and Receiver Operating Characteristic Area Under Curve. We found that using SMOTE techniques helped raise performance of the predictive models. Also, random forest (RF) had the best performance based on Matthews correlation coefficient (0.99), F-measure (0.99), Precision-Recall curve (1.00) and Receiver Operating Characteristic Area Under Curve (0.99).Conclusion: This discovery has the potential to impact on clinical practice, when health workers aim at classifying diagnosis result of disease at its early stage.

Performance Comparison of Data Sampling Techniques to Handle Imbalanced Class on Prediction of Compound-Protein Interaction

The prediction of Compound-Protein Interactions (CPI) is an essential step in the drug-target analysis for developing new drugs as well as for drug repositioning. One challenging issue in this field is that commonly there are more numbers of non-interacting compound-protein pairs than interacting pairs. This problem causes bias, which may degrade the prediction of CPI. Besides, currently, there is not much research on CPI prediction that compares data sampling techniques to handle the class imbalance problem. To address this issue, we compare four data sampling techniques, namely Random Under-sampling (RUS), Combination of Over-Under-sampling (COUS), Synthetic Minority Over-sampling Technique (SMOTE), and Tomek Link (T-Link). The benchmark CPI data: Nuclear Receptor and G-Protein Coupled Receptor (GPCR) are used to test these techniques. Area Under Curve (AUC) applied to evaluate the CPI prediction performance of each technique. Results show that the AUC values for RUS, COUS, SMOTE, and T-Link are 0.75, 0.77, 0.85 and 0.79 respectively on Nuclear Receptor data and 0.70, 0.85, 0.91 and 0.72 respectively on GPCR data. These results indicate that SMOTE has the highest AUC values. Furthermore, we found that the SMOTE technique is more capable of handling class imbalance problems on CPI prediction compared to the remaining three other techniques.

An Experimental Assessment of Random Forest Classification Performance Improvisation with Sampling and Stage Wise Success Rate Calculation

Imbalanced data classification with Random Forest Classification (RFC) technique has gained huge prominence in today’s application era. Data imbalance between practical applications relates to either binary class imbalance or multiclass imbalance. Binary class imbalance constitutes one of the classes with majority data samples and other contains minor number of data samples. In case of multiclass there are two categories of multiclass imbalanced dataset as Multiclass Minority Imbalanced Class (MMinIC) and Multiclass Majority Imbalanced Class (MMajIC). Classification performance leans towards degradation for MMajIC than MMinIC due to major imbalance rate severity. In this paper, the study investigates the influence of RFC classification analysis method on binary and multiclass sample imbalanced datasets. The analytical study of RFC incorporates with measurement of classification accuracy with performance metrics as True Positive (TP) Rate, False Positive (FP) Rate, Precision (Pre), Recall (Rec) F-Measure, Operating Characteristics of Receiver (ROC) Area, Matthews Correlation Coefficient (MCC), Probabilistic Relevance Classification (PRC) area with respect to numerous classes in refereed dataset. This paper focuses the reduction of the negative influence of imbalanced data with the use of Synthetic Minority Oversampling Technique (SMOTE). Experimental analysis carried out with the use of Knowledge Extraction Evolutionary Learning (KEEL) imbalanced data learning repository incorporating RFC classification with SMOTE technique. It also deals with RFC model construction with stage wise success rate calculation in training and testing partition and its impact on accuracy. The incorporates with error analysis report of incorrectly classified instances. Experimental results of the study indicate that imbalanced data have significant impact on classification accuracy and RFC outperforms with SMOTE.

A novel ensemble classification model based on neural networks and a classifier optimisation technique for imbalanced credit risk evaluation

Significant research has been performed on credit risk evaluation, with many machine learning and data mining techniques being employed for financial decision-making. The back propagation (BP) neural network has been a popular choice for credit risk evaluation problems, but many studies have found classifier ensembles to be superior to single classifiers. In this paper, a novel ensemble model based on the synthetic minority over-sampling technique (SMOTE) and a classifier optimisation technique is proposed for personal credit risk evaluation. To mitigate the negative effects of imbalanced datasets on the performance of the credit evaluation model, the SMOTE technique is used to rebalance the target training dataset. The particle swarm optimisation (PSO) algorithm is employed to search for the best-connected weights and deviations in the BP neural networks. Based on the optimised BP neural network classifiers, an ensemble model is developed that combines the AdaBoost approach with the base classifiers. To ensure that the proposed model provides accurate and stable performance, we thoroughly explore and discuss the optimal parameters for the ensemble classification model. Finally, the proposed ensemble model is tested on German and Australian real-world imbalanced datasets. The results demonstrate that this model is more effective at processing credit data problems compared to the other classification models examined in this study.

The Impact of Oversampling with SMOTE on the Performance of 3 Classifiers in Prediction of Type 2 Diabetes.

To evaluate the impact of the synthetic minority oversampling technique (SMOTE) on the performance of probabilistic neural network (PNN), naïve Bayes (NB), and decision tree (DT) classifiers for predicting diabetes in a prospective cohort of the Tehran Lipid and Glucose Study (TLGS). . Data of the 6647 nondiabetic participants, aged 20 years or older with more than 10 years of follow-up, were used to develop prediction models based on 21 common risk factors. The minority class in the training dataset was oversampled using the SMOTE technique, at 100%, 200%, 300%, 400%, 500%, 600%, and 700% of its original size. The original and the oversampled training datasets were used to establish the classification models. Accuracy, sensitivity, specificity, precision, F-measure, and Youden's index were used to evaluated the performance of classifiers in the test dataset. To compare the performance of the 3 classification models, we used the ROC convex hull (ROCCH). Oversampling the minority class at 700% (completely balanced) increased the sensitivity of the PNN, DT, and NB by 64%, 51%, and 5%, respectively, but decreased the accuracy and specificity of the 3 classification methods. NB had the best Youden's index before and after oversampling. The ROCCH showed that PNN is suboptimal for any class and cost conditions. To determine a classifier with a machine learning algorithm like the PNN and DT, class skew in data should be considered. The NB and DT were optimal classifiers in a prediction task in an imbalanced medical database.