Feature Selection Algorithm Research Articles

An interpretable model for large-scale smart contract vulnerability detection

Smart contracts hold billions of dollars in digital currency, and their security vulnerabilities have drawn a lot of attention in recent years. Traditional methods for detecting smart contract vulnerabilities rely primarily on symbol execution, which makes them time-consuming with high false positive rates. Recently, deep learning approaches have alleviated these issues but still face several major limitations, such as lack of interpretability and susceptibility to evasion techniques. In this paper, we propose a feature selection method for uplifting modeling. The fundamental concept of this method is a feature selection algorithm, utilizing interpretation outcomes to select critical features, thereby reducing the scales of features. The learning process could be accelerated significantly because of the reduction of the feature size. The experiment shows that our proposed model performs well in six types of vulnerability detection. The accuracy of each type is higher than 93% and the average detection time of each smart contract is less than 1 ms. Notably, through our proposed feature selection algorithm, the training time of each type of vulnerability is reduced by nearly 80% compared with that of its original.

A two-stage clonal selection algorithm for local feature selection on high-dimensional data

Various evolutionary computation algorithms have shown their excellent performance for high-dimensional feature selection (FS). However, most of current FS methods choose a global feature subset and ignore the correlations between feature subsets and sample subspaces, which limits the performance of methods in the sample space with different probability distributions. To solve the issue, we propose a two-stage clonal selection algorithm for filter-based local feature selection (TSCSA-LFS) that integrates symmetric uncertainty and a discrete clonal selection algorithm with three contributions. First, unlike conventional feature selection methods, TSCSA-LFS introduces local sample behaviors and assigns subsets of features for different sample regions. Second, an improved discrete clonal selection algorithm is developed for searching relevant features, which contains mutual information-based individual initialization, a differential evolution-based mutation strategy pool and a local search technique. Third, a two-part antibody representation is employed for automatical adjustment of the weight-related parameter. Our method shows the promising experimental results compared with well-known global FS and clonal selection-based local FS methods on fourteen high-dimensional datasets. For instance, our method can obviously outperform local FS, filter-based and hybrid methods on at least nine datasets

Enhancing Medical Image Classification with an Advanced Feature Selection Algorithm: A Novel Approach to Improving the Cuckoo Search Algorithm by Incorporating Caputo Fractional Order.

Glaucoma is a chronic eye condition that seriously impairs vision and requires early diagnosis and treatment. Automated detection techniques are essential for obtaining a timely diagnosis. In this paper, we propose a novel method for feature selection that integrates the cuckoo search algorithm with Caputo fractional order (CFO-CS) to enhance the performance of glaucoma classification. However, when using the infinite series, the Caputo definition has memory length truncation issues. Therefore, we suggest a fixed memory step and an adjustable term count for optimization. We conducted experiments integrating various feature extraction techniques, including histograms of oriented gradients (HOGs), local binary patterns (LBPs), and deep features from MobileNet and VGG19, to create a unified vector. We evaluate the informative features selected from the proposed method using the k-nearest neighbor. Furthermore, we use data augmentation to enhance the diversity and quantity of the training set. The proposed method enhances convergence speed and the attainment of optimal solutions during training. The results demonstrate superior performance on the test set, achieving 92.62% accuracy, 94.70% precision, 93.52% F1-Score, 92.98% specificity, 92.36% sensitivity, and 85.00% Matthew's correlation coefficient. The results confirm the efficiency of the proposed method, rendering it a generalizable and applicable technique in ophthalmology.

Pioneering Data Processing for Convolutional Neural Networks to Enhance the Diagnostic Accuracy of Traditional Chinese Medicine Pulse Diagnosis for Diabetes.

Traditional Chinese medicine (TCM) has relied on pulse diagnosis as a cornerstone of healthcare assessment for thousands of years. Despite its long history and widespread use, TCM pulse diagnosis has faced challenges in terms of diagnostic accuracy and consistency due to its dependence on subjective interpretation and theoretical analysis. This study introduces an approach to enhance the accuracy of TCM pulse diagnosis for diabetes by leveraging the power of deep learning algorithms, specifically LeNet and ResNet models, for pulse waveform analysis. LeNet and ResNet models were applied to analyze TCM pulse waveforms using a diverse dataset comprising both healthy individuals and patients with diabetes. The integration of these advanced algorithms with modern TCM pulse measurement instruments shows great promise in reducing practitioner-dependent variability and improving the reliability of diagnoses. This research bridges the gap between ancient wisdom and cutting-edge technology in healthcare. LeNet-F, incorporating special feature extraction of a pulse based on TMC, showed improved training and test accuracies (73% and 67%, respectively, compared with LeNet's 70% and 65%). Moreover, ResNet models consistently outperformed LeNet, with ResNet18-F achieving the highest accuracy (82%) in training and 74% in testing. The advanced preprocessing techniques and additional features contribute significantly to ResNet18-F's superior performance, indicating the importance of feature engineering strategies. Furthermore, the study identifies potential avenues for future research, including optimizing preprocessing techniques to handle pulse waveform variations and noise levels, integrating additional time-frequency domain features, developing domain-specific feature selection algorithms, and expanding the scope to other diseases. These advancements aim to refine traditional Chinese medicine pulse diagnosis, enhancing its accuracy and reliability while integrating it into modern technology for more effective healthcare approaches.

Optimized long short-term memory with rough set for sustainable forecasting renewable energy generation

Research and development in the field of renewable energy is receiving more attention as a result of the growing demand for clean, sustainable energy. This paper proposes a model for forecasting renewable energy generation. The proposed model consists of three main phases: data preparation, feature selection-based rough set and nutcracker optimization algorithm (NOA), and data classification and cross-validation. First, the missing values are tackled using the mean method. Then, data normalization and data shuffling are applied in the data preparation phase. In the second phase, a new feature selection algorithm is proposed based on rough set theory and NOA, namely RSNOA. The proposed RSNOA is based on adopting the rough set method as the fitness function during the searching mechanism to find the optimal feature subset. Finally, a custom long-short-term memory architecture with the k-fold cross-validation method is utilized in the last phase. The experimental results revealed that the proposed model is very competitive. It is achieved with 4.2113 root mean square error, 0.96 R², 2.835 mean absolute error, and 4.6349 mean absolute percentage error. The findings also show that the proposed model has great promise as a useful tool for accurately forecasting renewable energy generation across various sources.

Detection of chronic kidney disease using binary whale optimization algorithm

Chronic kidney disease (CKD), a medical illness, is characterized by a steady deterioration in kidney function. A disease's ability to be prevented and effectively significantly treated depends on early diagnosis. The addition of filter feature selection to the machine learning algorithm has been done to detect CKD. However, the quality of its feature subset is not optimal. Wrapper feature selection can improve the quality of these feature subsets. Therefore, we proposed wrapper feature selection and binary whale optimization algorithm (BWOA) to enhance the accuracy of early CKD detection. We also make data improvements to improve accuracy, namely the preprocessing process with the median and modus techniques. We used a public dataset of 250 medical records of kidney sufferers and 150 completely healthy people. There are 24 features in this dataset. The test results showed that adding BWOA feature selection can increase accuracy. The proposed method produced an accuracy of 100%. Further research on these methods can be used to develop expert systems for early detection of CKD.

Feature selection techniques for microarray dataset: a review

For many researchers working on feature selection techniques, finding an appropriate feature from the microarray dataset has turned into a bottleneck. Researchers often create feature selection approaches and algorithms with the goal of improving accuracy in microarray datasets. The main goal of this study is to present a variety of contemporary feature selection techniques, such as Filter, Wrapper, and Embedded methods proposed for microarray datasets to work on multi-class classification problems and different approaches to enhance the performance of learning algorithms, to address the imbalance issue in the data set, and to support research efforts on microarray dataset. This study is based on Critical Review Questions (CRQ) constructed using feature election methods described in the review methodology and applied to a microarray dataset. We discussed the analysed findings and future prospects of feature selection strategies for multi-class classification issues using microarray datasets, as well as prospective ways to speed up computing environment

Using machine learning approach based on gene expression profiling to predict the risk of bone metastasis in lung cancer.

e20522 Background: Bone metastasis (BM) occurs in about 30% of patients with advanced lung cancer, and the prognosis of BM patients is very poor, with a median survival of 6-10 months. Therefore, we attempted to use machine learning methods based on gene expression profiles to construct a model to predict BM. Methods: Gene Expression Omnibus (GEO) database were searched, then these datasets (GSE10096, GSE29367, GSE29391, GSE76194, GSE175601) were included for analysis. Batch effects were removed using the ComBat function from the R package sva (v3.44.0). edgeR was used for differential expression analysis, and genes that had an absolute log2 fold change > 1 and adjusted p-value < 0.05, were selected as differentially expressed genes (DEGs) for modeling. Random forest (RF) modeling was performed using the R package random Forest (v4.7-1.1), and support vector machine (SVM) modeling and SVM Recursive Feature Elimination (SVM-RFE) were performed using the R package e1071 (v1.7-13). Survival curves were generated using the GEPIA2 R packages. Models performance was evaluated using the area under the receiver operating characteristics curve (AUC). Results: A total of 48 qualified cases were subjected to analysis, including 22 BM and 26 non-BM patients. Differential expression analysis revealed that a total of 327 DEGs were identified in the BM group. We divided the sample dataset into a training set and a test set. In the first approach, we attempted to use all 327 DEGs features to train the models. Among the evaluated models, the RF algorithm performed best with an AUC of 0.753, while the SVM algorithm had an AUC of 0.938. The second approach, we applied the SVM-RFE algorithm for feature selection, which ranked the features according to their importance, and the top 10 genes/features (TMC5, S100A4, EEF1A2, CLDN1, RALY, MYH10, MAGEC2, ALDH1A1, PI3, SALL1) were used to build a classification model (in the test set, sensitivity was100%, specificity: 81.8%, AUC:0.909). In addition, TCGA survival analysis of 10 genes was performed and found that the expression levels of five genes, including ALDH1A1, MYH10, RALY, CLDN1 and MAGEC2, were significantly associated with prognosis (p < 0.1). Conclusions: Our study suggests that we have identified ten key genes associated with BM of lung cancer which could be potential therapeutic targets.

Hybrid machine learning for stock price prediction in the Moroccan banking sector

Analyzing historical stock market data using machine-learning techniques is crucial for data scientists and researchers to optimize stock price prediction models. This study uses machine learning regression algorithms and feature selection methods to optimize a simulated stock price prediction model using real historical data from Bank of Africa, a Moroccan bank. The approach compares multiple supervised regression algorithms, such as linear regression, extreme gradient boosting, ordinary least squared, random forest regressor, a linear least-squares L2-regularized, epsilon-support vector regression, and linear support vector regression. Each of these algorithms is associated with different feature selection algorithms to improve the performance of the prediction model. The analysis results revealed that hybridizing algorithms between the highest score percentiles, univariate linear regression, and linear support vector regression perform better according to the root mean squared error and R2-Score measures. This approach overcomes the problems associated with high-dimensional data by reducing the number of features and improving prediction accuracy.

Optimizing oil-source correlation analysis using support vector machines and sensory attention networks

Oil source correlation can be used to identify the origin of crude oil by linking crude oil to source rocks; however, the manual methods, which are limited by the sample or parameter quantity or imbalanced datasets, are facing uncertainties. Although the existing multivariate statistical techniques can alleviate this problem, they are facing difficulties in processing imbalanced datasets and quantifying source beds. Therefore, a novel oil-source correlation analysis model called SVM-SelectKBest combining a support vector machine (SVM) with a feature selection algorithm to mitigate the common issue of dataset imbalance in oil-source correlations is proposed in this paper. The SVM-SelectKBest offers advantages over normal SVM by dynamically selecting the most relevant features and fine-tuning model parameters to achieve higher accuracy and better generalizability in complex datasets. SVM compensates for class imbalances by heavily penalizing the misclassification of the minority class, and SelectKBest streamlines the feature set to enhance SVM's effectiveness on critical variables. Furthermore, a shallow neural network (SensoryAttentionNet) is introduced into the proposed model to address the issue of quantifying the source bed proportions in crude oil. The result show that SVM-SelectKBest has better performance in identifying key geochemical parameters and discriminating oil source correlation, its accuracy in unbalanced datasets is improved by near 40% compared to SVM. The model obtains 25 key geochemical parameters such as C17 n-heptadecane, Pr pristane, and C18 n-octadecane, it achieves F1 scores of 1.0 on the training, validation, and test sets. SensoryAttentionNet also performs robustly, with a low variance of 0.05 between its predicted and actual values. All the results indicate the effectiveness of the proposed method in dealing with the imbalance problem in oil-source source correlation datasets and in determining the proportional contribution of source beds in crude oil.

Automated feature selection using improved migrating birds optimization for enhanced medical diagnosis

The feature selection task is a crucial phase in data analysis, aiming to identify a minimized set of relevant features for the target class, thereby eliminating irrelevant and redundant attributes used for model training. While population-based feature selection approaches offer prominent solutions for classification performance, their computational time can be prohibitive. To mitigate delays and optimize resource utilization, this study adopts machine learning operations (MLOps). MLOps involves the seamless transition of experimental Machine Learning models into production, serving them to end users and automating the feature selection phase. This paper introduces a novel feature selection method based on improved migrating bird optimization and its automated variant integrated into MLOps. Experiments conducted on six medical datasets validate the effectiveness of our proposed feature selection method in improving the outcomes of medical diagnosis systems. The results showcase satisfactory performance in terms of classification compared to concurrent feature selection algorithms.

Improving dynamic prognostication of patients with hepatocellular carcinoma by node fusion algorithm equipped survival path mapping.

e13594 Background: Patients with hepatocellular carcinoma (HCC) always require routine surveillance and repeated treatment. Our previous work (1) developed a machine learning methodology called “Survival Path Mapping”, which converted the time series data into a cascading survival map to facilitate dynamic prognosis prediction and treatment planning. However, in previous work, the exponential growth of paths raised challenges of exhausting limited cohort and over-fitting risk. Novel algorithm enhancing data utilization in the Survival Path Mapping was in urgent need. Methods: A node fusion algorithm was designed. First, feature selection algorithm bifurcated each parent-nodes into temporary child-nodes, which constitutes temporary child-nodes set. Then, clustering was performed within temporary child-nodes set according to key features of each child-node. Finally, in each cluster, multiple child-nodes would be merged into one if there was no significant difference in survival. For nodes with bifurcation, a treatment recommendation algorithm was developed. A total of 19446 patients with newly diagnosed primary HCC at Sun Yat-sen University Cancer Center (SYSUCC) between 2008 and 2023 were retrospectively included as internal cohort (divided into internal train and holdout validation cohort); meanwhile 1135 patients from five hospitals were included as multicenter external validation cohort. Dedicated natural language processing program was developed to extract structured features from medical imaging reports. The time series data of patients were converted into data of time slices, with an interval of 3 months. Results: The original survival path mapping (raw-SP) consists of 463 different nodes, while the node fusion survival path mapping (fusion-SP) consisted of 84 nodes (81.8% decrease from raw-SP). The 3-year c-index of fusion-SP were 0.745 (95% CI 0.717-0.772), significantly better (p<0.001) than that of raw-SP: 0.725 (95% CI 0.696-0.755), in the internal holdout validation. Both raw-SP and fusion-SP significantly (p < 0.001) outperformed conventional BCLC and CNLC staging. In the multicenter external validation, the fusion-SP also demonstrated better performance than raw-SP and BCLC staging system. The recommended therapy calculated based fusion-SP system for patients receiving second-line or subsequent treatment had superior or non-inferior efficacy compared to BCLC guideline’s recommendation. Conclusions: The proposed node fusion algorithm solved exponential growth of paths in previous survival path algorithm, improving the accuracy and external validity of dynamic prognosis prediction for HCC patients. The fusion-SP model also retained its transparency and could be used to guide dynamic treatment planning, with its effectiveness in treatment guidance evaluated in prospective trials in the future. 1. Nat Commun 9, 2230; 2018.

Prediction of failures in sewer networks using various machine learning classifiers

ABSTRACT In sewer networks, failure prediction plays a significant role in operation and maintenance plans of wastewater utilities. This study aims to determine the effective variables on the failures by using feature selection algorithms (FS) and achieve maximum model accuracy with minimum variables. Also, four scenarios based on the suggested FS algorithms were developed. In these scenarios, the best prediction models were investigated using machine learning classifiers (ML) such as neural network classifier (NNC), gradient boosting machine (GBM), random forest (FR), and hybrid model (HM). The classification performance of ML models was evaluated using accuracy, precision, F1_score, and receiver operating characteristics (ROC) curve. The model accuracies ranging from 0.99 for accuracy to 1 for the ROC curve were achieved through ML algorithms. In conclusion, the ML algorithms suggested in this study may be a decision support tool for wastewater utilities in prioritizing the replacement, maintenance, and inspection of sewer pipes.

Risk prediction of coal mine rock burst based on machine learning and feature selection algorithm

ABSTRACT With the intensification of coal mining in China, rock burst hazards have become increasingly frequent and serious. To comprehensively investigate the accident indicators during mining activities, identify the key triggering factors, and better prevent rockburst accidents, this study establishes a rockburst risk prediction index system containing 16 indicators in four subsystems: geological structure, mining technology, safety management, and employee situation. By using 118 cases as sample data, this study proposes to apply four feature selection methods (Boruta, PC, RF, BE) to reveal the relationship between these indicators and select the most salient indicators for risk prediction. Then four machine learning algorithms, namely Random Forest (RF), Gradient Boosting Machine (GBM), SVM, and KNN are used to compare the effects of different feature selection methods on the prediction accuracy of risk level. The results show that Boruta performs best as a feature selection method for RF, GBM, and KNN, and the Boruta-RF model has better predictive performance than the others.

Leveraging Various Feature Selection Methods for Churn Prediction Using Various Machine Learning Algorithms

This study aims to examine the effect of customer experience on customer retention at DQLab Telco, using machine learning techniques to predict customer churn. The study uses a dataset of 6590 customers of DQLab Telco, which contains various features related to their service usage and satisfaction. The data includes various features such as gender, tenure, phone service, internet service, monthly charges, and total charges. These features represent the demographic and service usage information of the customers. The study applies several feature selection methods, such as ANOVA, Recursive Feature Elimination, Feature Importance, and Pearson Correlation, to select the most relevant features for churn prediction. The study also compares three machine learning algorithms, namely Logistic Regression, Random Forest, and Gradient Boosting, to build and evaluate the prediction models. This study finds that Logistic Regression without feature selection achieves the highest accuracy of 79.47%, while Random Forest with Feature Importance and Gradient Boosting with Recursive Feature Elimination achieve accuracy of 77.60% and 79.86%, respectively. The study also identifies the features influencing customer churn most, such as monthly charges, tenure, partner, senior citizen, internet service, paperless billing, and TV streaming. The study provides valuable insights for DQLab Telco in developing customer churn reduction strategies based on predictive models and influential features. The study also suggests that feature selection and machine learning algorithms play a vital role in improving the accuracy of churn prediction and should be customized according to the data context.

Adware Detection using LSTM and Comparison with other ML Models

Abstract: In recent years, the internet has become the predominant communication medium, paralleled by a surge in cyber threats where attackers actively exploit vulnerabilities to obtain sensitive information. Among the various malware tactics employed, Adware poses a significant challenge. Addressing this concern, cyber security specialists leverage Machine Learning (ML) techniques, with this paper proposing a novel Long Short-Term Memory (LSTM) algorithm for adware detection. The research employs a holistic methodology involving diverse data pre-processing techniques, feature selection, and ML algorithms to effectively identify adware samples within the dataset. A comparative analysis of ML classifiers, including Random Forest (RF), k - Nearest Neighbors (k-NN), Decision Tree (DT), and Logistic Regression (LR), reveals optimal detection accuracies of 98.66%, 98.10%, and 98.05% for DT and k-NN, respectively. These findings underscore the efficacy of the proposed approach in fortifying cyber security against adware threats.

Application of Salp Swarm Algorithm and Extended Repository Feature Selection Method in Bearing Fault Diagnosis

Motor fault diagnosis is an important task in the operational monitoring of electrical machines in manufacturing. This study proposes an effective bearing fault diagnosis model for electrical machinery based on machine learning techniques. The proposed model is a combination of three processes: feature extraction of signals collected from the motor based on multi-resolution analysis, fast Fourier transform, and envelope analysis. Next, redundant or irrelevant features are removed using the feature selection technique. A binary salps swarm algorithm combined with an extended repository is the proposed method to remove unnecessary features. As a result, an optimal feature subset is obtained to improve the performance of the classification model. Finally, two classifiers, k-nearest neighbor and support vector machine, are used to classify the fault of the electric motor. There are four input datasets used to evaluate the model performance, and UCI is the benchmark dataset to verify the effectiveness of the proposed feature selection technique. The remaining three datasets include the bearing dataset collected from experiments, with an average classification accuracy of 99.9%, as well as Case Western Reserve University (CWRU) and Machinery Failure Prevention Technology (MFPT), which are public datasets with average classification accuracies of 99.6% and 98.98%, respectively. The experimental results show that this method is more effective in diagnosing bearing faults than other traditional methods and prove its robustness.

Swing Trend Prediction of Main Guide Bearing in Hydropower Units Based on MFS-DCGNN.

Hydropower units are the core equipment of hydropower stations, and research on the fault prediction and health management of these units can help improve their safety, stability, and the level of reliable operation and can effectively reduce costs. Therefore, it is necessary to predict the swing trend of these units. Firstly, this study considers the influence of various factors, such as electrical, mechanical, and hydraulic swing factors, on the swing signal of the main guide bearing y-axis. Before swing trend prediction, the multi-index feature selection algorithm is used to obtain suitable state variables, and the low-dimensional effective feature subset is obtained using the Pearson correlation coefficient and distance correlation coefficient algorithms. Secondly, the dilated convolution graph neural network (DCGNN) algorithm, with a dilated convolution graph, is used to predict the swing trend of the main guide bearing. Existing GNN methods rely heavily on predefined graph structures for prediction. The DCGNN algorithm can solve the problem of spatial dependence between variables without defining the graph structure and provides the adjacency matrix of the graph learning layer simulation, avoiding the over-smoothing problem often seen in graph convolutional networks; furthermore, it effectively improves the prediction accuracy. The experimental results showed that, compared with the RNN-GRU, LSTNet, and TAP-LSTM algorithms, the MAEs of the DCGNN algorithm decreased by 6.05%, 6.32%, and 3.04%; the RMSEs decreased by 9.21%, 9.01%, and 2.83%; and the CORR values increased by 0.63%, 1.05%, and 0.37%, respectively. Thus, the prediction accuracy was effectively improved.

Online System for Monitoring the Degree of Fermentation of Oolong Tea Using Integrated Visible-Near-Infrared Spectroscopy and Image-Processing Technologies.

During the fermentation process of Oolong tea, significant changes occur in both its external characteristics and its internal components. This study aims to determine the fermentation degree of Oolong tea using visible-near-infrared spectroscopy (vis-VIS-NIR) and image processing. The preprocessed vis-VIS-NIR spectral data are fused with image features after sequential projection algorithm (SPA) feature selection. Subsequently, traditional machine learning and deep learning classification models are compared, with the support vector machine (SVM) and convolutional neural network (CNN) models yielding the highest prediction rates among traditional machine learning models and deep learning models with 97.14% and 95.15% in the prediction set, respectively. The results indicate that VIS-NIR combined with image processing possesses the capability for rapid non-destructive online determination of the fermentation degree of Oolong tea. Additionally, the predictive rate of traditional machine learning models exceeds that of deep learning models in this study. This study provides a theoretical basis for the fermentation of Oolong tea.

Dynamic Nondestructive Detection Models of Apple Quality in Critical Harvest Period Based on Near-Infrared Spectroscopy and Intelligent Algorithms.

Apples are usually bagged during the growing process, which can effectively improve the quality. Establishing an in situ nondestructive testing model for in-tree apples is very important for fruit companies in selecting raw apple materials for valuation. Low-maturity apples and high-maturity apples were acquired separately by a handheld tester for the internal quality assessment of apples developed by our group, and the effects of the two maturity levels on the soluble solids content (SSC) detection of apples were compared. Four feature selection algorithms, like ant colony optimization (ACO), were used to reduce the spectral complexity and improve the apple SSC detection accuracy. The comparison showed that the diffuse reflectance spectra of high-maturity apples better reflected the internal SSC information of the apples. The diffuse reflectance spectra of the high-maturity apples combined with the ACO algorithm achieved the best results for SSC prediction, with a prediction correlation coefficient (Rp) of 0.88, a root mean square error of prediction (RMSEP) of 0.5678 °Brix, and a residual prediction deviation (RPD) value of 2.466. Additionally, the fruit maturity was predicted using PLS-LDA based on color data, achieveing accuracies of 99.03% and 99.35% for low- and high-maturity fruits, respectively. These results suggest that in-tree apple in situ detection has great potential to enable improved robustness and accuracy in modeling apple quality.