Recursive Feature Elimination Technique Research Articles

Predictive analysis of cardiovascular disease using gradient boosting based learning and recursive feature elimination technique

• The RFE-GB method is proposed in this study for obtaining accurate heart disease prediction. • The health records of patients with CVD characteristics were examined in order to assess the outcomes. • According to the findings, RFE has the best accuracy (88.8 percent). • With an area under the curve of 0.85, the RFE-GB algorithm outperformed other methods. Heart disease is one of the most frequent chronic ailments people suffer. Early identification can lower death rates by avoiding or lowering cardiovascular disease (CVD) severity. For detecting risk indicators, machine learning algorithms are a potential way. To acquire accurate cardiac disease prediction, this work introduces a recursive feature elimination-based gradient boosting (RFE-GB) approach. The outcomes were evaluated using the patients' health records, including crucial CVD characteristics. The prediction model was built using many additional machine learning approaches, and the results were compared to the suggested model. The combined recursive feature removal and gradient boosting approach deliver the maximum accuracy, according to the findings of this suggested model (88.8 %). Furthermore, the presented RFE-GB method was determined to be superior and had acquired a significant gain over previous strategies, with an area under the curve of 0.84. The developed RFE-GB method will thus be a useful model for predicting and treating CVD.

Explainable Artificial Intelligence (XAI) Model for the Diagnosis of Urinary Tract Infections in Emergency Care Patients

Significance of machine learning (ML), deep learning (DL) techniques and the availability of Electronic Health Records (EHR) has motivated the need of automated diagnosis system. Furthermore, this development has transformed the health care systems. Recently, several ML and DL models has been proposed for various diseases and has shown the significant outcomes as well. Unfortunately, Urinary tract infections (UTI) is among the minor diseases that is not investigated a lot interms of diagnosing using computation intelligence techniques. However, these models lack the reliability due to the black box nature of the highly complex logic model. Therefore, we attempt to develop an interpretable deep learning (DL) model for the diagnosis of UTI using the dataset of emergency department (ED) patients from UK. Several sets of experiments were conducted using complete dataset, reduced attribute set identified using recursive feature elimination (RFE) and using the attributes identified by the baseline study. The proposed DL model has improved the baseline study accuracy from 0.875 to 0.9275 for 184 feature and 0.859 to 0.943 for the reduced feature. Furthermore, the model has outperformed interms of sensitivity and specificity as well. Due to the data imbalance positive predicted value (PPV), negative predicted value (NPV) and Youden Index was also used for evaluating the performance of the model. The proposed DL model has achieved the highest outcome using 18 attributes selected with RFE technique. The proposed model will produce reliability in the diagnosis made by the model and provide confidence to the doctors to adopt the system in the real life.

Research on imbalance machine learning methods for MRT_1WI soft tissue sarcoma data

BackgroundSoft tissue sarcoma is a rare and highly heterogeneous tumor in clinical practice. Pathological grading of the soft tissue sarcoma is a key factor in patient prognosis and treatment planning while the clinical data of soft tissue sarcoma are imbalanced. In this paper, we propose an effective solution to find the optimal imbalance machine learning model for predicting the classification of soft tissue sarcoma data.MethodsIn this paper, a large number of features are first obtained based on T_1WI images using the radiomics methods.Then, we explore the methods of feature selection, sampling and classification, get 17 imbalance machine learning models based on the above features and performed extensive experiments to classify imbalanced soft tissue sarcoma data. Meanwhile, we used another dataset splitting method as well, which could improve the classification performance and verify the validity of the models.ResultsThe experimental results show that the combination of extremely randomized trees (ERT) classification algorithm using SMOTETomek and the recursive feature elimination technique (RFE) performs best compared to other methods. The accuracy of RFE+STT+ERT is 81.57% , which is close to the accuracy of biopsy, and the accuracy is 95.69% when using another dataset splitting method.ConclusionPreoperative predicting pathological grade of soft tissue sarcoma in an accurate and noninvasive manner is essential. Our proposed machine learning method (RFE+STT+ERT) can make a positive contribution to solving the imbalanced data classification problem, which can favorably support the development of personalized treatment plans for soft tissue sarcoma patients.

An anomaly-based intrusion detection system using recursive feature elimination technique for improved attack detection

The explosive growth of Internet and technologies has resulted in better life-changing services to the society. This drastic change has invariably kept the users connected to the Internet uninterruptedly. The adversaries might exploit this “always-on” Internet connection and would jeopardize the network infrastructure through malicious activities using the networks' hardware and software vulnerabilities. However, the network traffic volume poses difficulties for the cyber analyst to monitor, detect, identify, and quickly respond to the network attacks. Therefore it is paramount to protect the integrity, availability, confidentiality of the network infrastructure, and its digital assets by deploying Intrusion Detection Systems. However, augmenting new techniques avoids overwhelming the analysts from a myriad of events. This research aims to analyze the traffic for the significant features that assist in reliable intrusion detection and minimize the resource requirements and computational complexity of the detection process. The proposed method uses Recursive Feature Elimination technique to rank features based on feature importance and Random Forest algorithm performs the attack classification. The experimental findings show these significant 21 features are useful in anomaly detection with an accuracy of 99.83% for binary attack detection, 99.69% for multiclass classification with an execution time of 20 seconds, and 39 seconds respectively on the NSL-KDD dataset.

Machine Learning Coronary Artery Disease Prediction Based on Imaging and Non-Imaging Data.

The prediction of obstructive atherosclerotic disease has significant clinical meaning for the decision making. In this study, a machine learning predictive model based on gradient boosting classifier is presented, aiming to identify the patients of high CAD risk and those of low CAD risk. The machine learning methodology includes five steps: the preprocessing of the input data, the class imbalance handling applying the Easy Ensemble algorithm, the recursive feature elimination technique implementation, the implementation of gradient boosting classifier, and finally the model evaluation, while the fine tuning of the presented model was implemented through a randomized search optimization of the model’s hyper-parameters over an internal 3-fold cross-validation. In total, 187 participants with suspicion of CAD previously underwent CTCA during EVINCI and ARTreat clinical studies and were prospectively included to undergo follow-up CTCA. The predictive model was trained using imaging data (geometrical and blood flow based) and non-imaging data. The overall predictive accuracy of the model was 0.81, using both imaging and non-imaging data. The innovative aspect of the proposed study is the combination of imaging-based data with the typical CAD risk factors to provide an integrated CAD risk-predictive model.

Analysis of Accuracy in Heart Disease Diagnosis System Using Decision Tree Classifier Over Logistic Regression Based on Recursive Feature Selection

The aim of the research paper is to find the accuracy in a better way using Decision Tree compared with Logistic Regression by using Recursive Feature Elimination Technique. Materials and Methods: In this study, there are two groups, namely Decision Tree and Logistic Regression. Accuracy was computed for the dataset with sample size of 40. The innovative method used is Recursive Feature Elimination. It is used to find the subset of features which gives more accuracy. Result: It was observed that the Decision Tree algorithm obtains accuracy of 85.5% and Logistic Regression with 83.4% of accuracy. Decision Tree appears to have better significance than Logistic Regression technique with the value of p<0.05. The significant value obtained from statistical analysis is 0.01. Conclusion: The result proves that Decision Tree classifier shows better accuracy than Logistic Regression classifier based on Recursive Feature Selection.

An Intelligent and Reliable Hyperparameter Optimization Machine Learning Model for Early Heart Disease Assessment Using Imperative Risk Attributes.

Heart disease is a severe disorder, which inflicts an adverse burden on all societies and leads to prolonged suffering and disability. We developed a risk evaluation model based on visible low-cost significant noninvasive attributes using hyperparameter optimization of machine learning techniques. The multiple set of risk attributes is selected and ranked by the recursive feature elimination technique. The assigned rank and value to each attribute are validated and approved by the choice of medical domain experts. The enhancements of applying specific optimized techniques like decision tree, k-nearest neighbor, random forest, and support vector machine to the risk attributes are tested. Experimental results show that the optimized random forest risk model outperforms other models with the highest sensitivity, specificity, precision, accuracy, AUROC score, and minimum misclassification rate. We simulate the results with the prevailing research; they show that it can do better than the existing risk assessment models with exceptional predictive accuracy. The model is applicable in rural areas where people lack an adequate supply of primary healthcare services and encounter barriers to benefit from integrated elementary healthcare advances for initial prediction. Although this research develops a low-cost risk evaluation model, additional research is needed to understand newly identified discoveries about the disease.

Saline soil moisture mapping using Sentinel-1A synthetic aperture radar data and machine learning algorithms in humid region of China's east coast

Accurate and timely information on soil moisture (SM) is crucial for better understanding the hydrological and ecological processes in humid saline regions. This study investigated the potential of using Sentinel-1A Synthetic Aperture Radar (SAR) imagery and Machine Learning Algorithms for SM mapping in China's east coast. This study used the recursive feature elimination (RFE) technique to select optimum SAR indices from the backscattering coefficients of the sampling date as well as the averaged backscattering coefficients of the neighbouring dates. The newly developed support vector regression (SVR) model provided more accurate SM estimation than that from the random forest regression (RFR) and multivariate linear regression (MLR) models. By using the leave-one-out cross-validation (LOOCV), the achieved coefficient of determination (R2), root mean square error (RMSE), and the relative RMSE (RRMSE) in SM estimation were 0.77, 1.82%, and 10.29%, respectively. The SVR model was able to map the SM effectively over a large area under the condition that the information of soil surface roughness, soil salinity, and land cover types was unknown. Results from this study demonstrate that high-temporal-resolution Sentinel-1A SAR, with its free-accessibility, offers a great source for the SVR model to provide frequent and accurate SM mapping over coastal saline regions.

Mapping of soil organic carbon using machine learning models: Combination of optical and radar remote sensing data

AbstractSoil organic C (SOC) plays an important role in soil quality. Thus, it is of great significance to know the spatial distribution characteristics of SOC. Environmental variables, such as natural predictors, remote sensing (RS) variables, and digital soil mapping approaches have been widely used in SOC prediction. However, it is still challenging to determine which methods and variables are effective to map SOC in farmland. In this study, we compare the performance of three machine learning models, including random forest, Cubist, and the support vector machine model for the prediction of SOC contents using RS variables, topographic variables, climate variables, soil property variables, and spatial position variables as environmental covariates. The recursive feature elimination (RFE) technique was used to select optimal variables. Of the 130 sampling sites of the cultivated layer that were collected in Shanggao county, 70% of the data were used for training and 30% for validation. Prediction accuracy and uncertainty was quantified using the bootstrap approach with 100 iterations. The performance of the model was verified by the mean absolute error, RMSE, R2, and Lin's concordance correlation coefficient (CCC); the SD was used as a measure of prediction uncertainty. Results showed that the Cubist model combined with the RFE method (Cubist_RFE) had a mean R2, RMSE, and SD of .84, 2.39 g kg–1, and 0.73 g kg–1, respectively, which showed the highest prediction accuracy and the lowest uncertainty. The soil property variables (soil alkaline hydrolysis N and pH) and RS variables (soil adjusted vegetation index, enhanced vegetation index, and band7 of Sentinel‐2) were the controlling factors of SOC prediction. Thus, the results indicate the potential of the Cubist model in SOC prediction, and the combination of the optical and radar RS data is a viable way for predicting SOC.

A Minimalist Method Toward Severity Assessment and Progression Monitoring of Obstructive Sleep Apnea on the Edge

Artificial Intelligence-enabled applications on edge devices have the potential to revolutionize disease detection and monitoring in future smart health (sHealth) systems. In this study, we investigated a minimalist approach for the severity classification, severity estimation, and progression monitoring of obstructive sleep apnea (OSA) in a home environment using wearables. We used the recursive feature elimination technique to select the best feature set of 70 features from a total of 200 features extracted from polysomnogram. We used a multi-layer perceptron model to investigate the performance of OSA severity classification with all the ranked features to a subset of features available from either Electroencephalography or Heart Rate Variability (HRV) and time duration of SpO2 level. The results indicate that using only computationally inexpensive features from HRV and SpO2, an area under the curve of 0.91 and an accuracy of 83.97% can be achieved for the severity classification of OSA. For estimation of the apnea-hypopnea index, the accuracy of RMSE = 4.6 and R-squared value = 0.71 have been achieved in the test set using only ranked HRV and SpO2 features. The Wilcoxon-signed-rank test indicates a significant change (p < 0.05) in the selected feature values for a progression in the disease over 2.5 years. The method has the potential for integration with edge computing for deployment on everyday wearables. This may facilitate the preliminary severity estimation, monitoring, and management of OSA patients and reduce associated healthcare costs as well as the prevalence of untreated OSA.

Optimal Triage for COVID-19 Patients Under Limited Health Care Resources With a Parsimonious Machine Learning Prediction Model and Threshold Optimization Using Discrete-Event Simulation: Development Study.

BackgroundThe COVID-19 pandemic has placed an unprecedented burden on health care systems.ObjectiveWe aimed to effectively triage COVID-19 patients within situations of limited data availability and explore optimal thresholds to minimize mortality rates while maintaining health care system capacity.MethodsA nationwide sample of 5601 patients confirmed with COVID-19 until April 2020 was retrospectively reviewed. Extreme gradient boosting (XGBoost) and logistic regression analysis were used to develop prediction models for the maximum clinical severity during hospitalization, classified according to the World Health Organization Ordinal Scale for Clinical Improvement (OSCI). The recursive feature elimination technique was used to evaluate the maintenance of model performance when clinical and laboratory variables were eliminated. Using populations based on hypothetical patient influx scenarios, discrete-event simulation was performed to find an optimal threshold within limited resource environments that minimizes mortality rates.ResultsThe cross-validated area under the receiver operating characteristic curve (AUROC) of the baseline XGBoost model that utilized all 37 variables was 0.965 for OSCI ≥6. Compared to the baseline model’s performance, the AUROC of the feature-eliminated model that utilized 17 variables was maintained at 0.963 with statistical insignificance. Optimal thresholds were found to minimize mortality rates in a hypothetical patient influx scenario. The benefit of utilizing an optimal triage threshold was clear, reducing mortality up to 18.1%, compared with the conventional Youden index.ConclusionsOur adaptive triage model and its threshold optimization capability revealed that COVID-19 management can be achieved via the cooperation of both the medical and health care management sectors for maximum treatment efficacy. The model is available online for clinical implementation.

MRI radiomics to differentiate between low grade glioma and glioblastoma peritumoral region.

The peritumoral region (PTR) of glioblastoma (GBM) appears as a T2W-hyperintensity and is composed of microscopic tumor and edema. Infiltrative low grade glioma (LGG) comprises tumor cells that seem similar to GBM PTR on MRI. The work here explored if a radiomics-based approach can distinguish between the two groups (tumor and edema versus tumor alone). Patients with GBM and LGG imaged using a 1.5T MRI were included in the study. Image data from cases of GBM PTR, and LGG were manually segmented guided by T2W hyperintensity. A set of 91 first-order and texture features were determined from each of T1W-contrast, and T2W-FLAIR, diffusion-weighted imaging sequences. Applying filtration techniques, a total of 3822 features were obtained. Different feature reduction techniques were employed, and a subsequent model was constructed using four machine learning classifiers. Leave-one-out cross-validation was used to assess classifier performance. The analysis included 42 GBM and 36 LGG. The best performance was obtained using AdaBoost classifier using all the features with a sensitivity, specificity, accuracy, and area of curve (AUC) of 91%, 86%, 89%, and 0.96, respectively. Amongst the feature selection techniques, the recursive feature elimination technique had the best results, with an AUC ranging from 0.87 to 0.92. Evaluation with the F-test resulted in the most consistent feature selection with 3 T1W-contrast texture features chosen in over 90% of instances. Quantitative analysis of conventional MRI sequences can effectively demarcate GBM PTR from LGG, which is otherwise indistinguishable on visual estimation.

Identifying Children at Readmission Risk: At-Admission versus Traditional At-Discharge Readmission Prediction Model.

The timing of 30-day pediatric readmissions is skewed with approximately 40% of the incidents occurring within the first week of hospital discharges. The skewed readmission time distribution coupled with delay in health information exchange among healthcare providers might offer a limited time to devise a comprehensive intervention plan. However, pediatric readmission studies are thus far limited to the development of the prediction model after hospital discharges. In this study, we proposed a novel pediatric readmission prediction model at the time of hospital admission which can improve the high-risk patient selection process. We also compared proposed models with the standard at-discharge readmission prediction model. Using the Hospital Cost and Utilization Project database, this prognostic study included pediatric hospital discharges in Florida from January 2016 through September 2017. Four machine learning algorithms—logistic regression with backward stepwise selection, decision tree, Support Vector machines (SVM) with the polynomial kernel, and Gradient Boosting—were developed for at-admission and at-discharge models using a recursive feature elimination technique with a repeated cross-validation process. The performance of the at-admission and at-discharge model was measured by the area under the curve. The performance of the at-admission model was comparable with the at-discharge model for all four algorithms. SVM with Polynomial Kernel algorithms outperformed all other algorithms for at-admission and at-discharge models. Important features associated with increased readmission risk varied widely across the type of prediction model and were mostly related to patients’ demographics, social determinates, clinical factors, and hospital characteristics. Proposed at-admission readmission risk decision support model could help hospitals and providers with additional time for intervention planning, particularly for those targeting social determinants of children’s overall health.

Digital Biomarkers of Physical Frailty and Frailty Phenotypes Using Sensor-Based Physical Activity and Machine Learning.

Remote monitoring of physical frailty is important to personalize care for slowing down the frailty process and/or for the healthy recovery of older adults following acute or chronic stressors. Taking the Fried frailty criteria as a reference to determine physical frailty and frailty phenotypes (slowness, weakness, exhaustion, inactivity), this study aimed to explore the benefit of machine learning to determine the least number of digital biomarkers of physical frailty measurable from a pendant sensor during activities of daily living. Two hundred and fifty-nine older adults were classified into robust or pre-frail/frail groups based on the physical frailty assessments by the Fried frailty criteria. All participants wore a pendant sensor at the sternum level for 48 h. Of seventeen sensor-derived features extracted from a pendant sensor, fourteen significant features were used for machine learning based on logistic regression modeling and a recursive feature elimination technique incorporating bootstrapping. The combination of percentage time standing, percentage time walking, walking cadence, and longest walking bout were identified as optimal digital biomarkers of physical frailty and frailty phenotypes. These findings suggest that a combination of sensor-measured exhaustion, inactivity, and speed have potential to screen and monitor people for physical frailty and frailty phenotypes.

Identification and Characterization of Senescence Phenotype in Lung Adenocarcinoma with High Drug Sensitivity

Lung adenocarcinoma (LUAD) is a major health problem and has poor prognosis. Heterogeneity is a central determinant of the treatment outcome, requiring identification of new subclasses of LUAD. Senescence has emerged as a crucial regulator of metastasis and drug response. Ionizing radiation- and doxorubicin-induced senescence-associated genes in lung fibroblasts were used in K-means clustering to identify high- and low-senescence (HS and LS) classes among The Cancer Genome Atlas- LUAD (TCGA-LUAD) patients. The LS group showed significantly poorer survival (P = 0.01) and greater activation of proliferative signaling pathways, proliferation, wound healing, and genetic aberrations (P < 0.05). The TP53 mutation rate was significantly greater in the HS group (P < 0.0001), explaining the phenotype. Also, genome-wide hypomethylation was significantly greater in the LS group than in the HS group. Interestingly, pathway analysis identified silencing of Wnt signaling in the HS group. The machine learning-based recursive feature elimination technique was used to identify a 20-gene senescence signature in TCGA-LUAD samples. The presence of a senescence phenotype with poor survival was validatedin an independent patient cohort and a cell-line cohort using unsupervised clustering of samplesbased ona 20-gene signature. On further analysis, HS cells were more resistant to drugs, particularly histone deacetylase inhibitors. Taken together, this study identified a novel subtype of LUAD with reduced Wnt signaling and high drug resistance.

An optimal intrusion detection system using recursive feature elimination and ensemble of classifiers

With the rapid growth and advancement in technology, we are becoming more and more dependent on the internet for our day-to-day work. As a result of this, we have become an easy target for the attackers. More the usage of the internet more we are vulnerable to threats. In this scenario, the need for anti-virus or some system that can help in detecting the threats or attacks is also growing rapidly. The answer to this problem is a system that when installed on a network shall be able to detect Intrusions of any sort. Such a system can be called an IDS (Intrusion Detection System). This system is deployed on any network to keep a track of the traffic and to monitor any kind of mutation or deviations from regular traffic patterns. Many kinds of research are still ongoing in this field to develop a system that has not only a better error detection rate but also can introduce preventive measures as soon as a threat is detected. This paper proposes a system for detecting the attacks using machine learning where the Recursive Feature Elimination (RFE) technique is applied when the irrelevant features are there in the dataset. This technique helps remove any sort of redundancy in the KDD CUP99 dataset which is a standard dataset for network security and intrusion detection. Then for the achieved set of features a confusion matrix is generated and the classification of records is done using decision tree, support vector machine, and an ensemble classifier random forest in the form of discriminant analysis. When compared with other methodologies our approach holds a good classification rate for all the classes of attacks of the KDD CUP 99 dataset.

Predicting the pathological status of mammographic microcalcifications through a radiomics approach

Objective The study aimed to develop a machine learning (ML)-coupled interpretable radiomics signature to predict the pathological status of non-palpable suspicious breast microcalcifications (MCs).Methods We enrolled 463 digital mammographical view images from 260 consecutive patients detected with non-palpable MCs and BI-RADS scored at 4 (training cohort, n = 428; independent testing cohort, n = 35) in the First Affiliated Hospital of Nanjing Medical University between September 2010 and January 2019. Subsequently, 837 textures and 9 shape features were subsequently extracted from each view and finally selected by an XGBoost-embedded recursive feature elimination technique (RFE), followed by four machine learning-based classifiers to build the radiomics signature.Results Ten radiomic features constituted a malignancy-related signature for breast MCs as logistic regression (LR) and support vector machine (SVM) yielded better positive predictive value (PPV)/sensitivity (SE), 0.904 (95% CI, 0.865–0.949)/0.946 (95% CI, 0.929–0.977) and 0.891 (95% CI, 0.822–0.939)/0.939 (95% CI, 0.907–0.973) respectively, outperforming their negative predictive value (NPV)/specificity (SP) from 10-fold cross-validation (10FCV) of the training cohort. The optimal prognostic model was obtained by SVM with an area under the curve (AUC) of 0.906 (95% CI, 0.834–0.969) and accuracy (ACC) 0.787 (95% CI, 0.680–0.855) from 10FCV against AUC 0.810 (95% CI, 0.760–0.960) and ACC 0.800 from the testing cohort.Conclusion The proposed radiomics signature dependens on a set of ML-based advanced computational algorithms and is expected to identify pathologically cancerous cases from mammographically undecipherable MCs and thus offer prospective clinical diagnostic guidance.

Age-gender specific prediction model for Parkinson’s severity assessment using gait biomarkers

Parkinson’s disease (PD) causes gait impairments resulting in tremor, balance instabilities, increased fall risk, and disability. The current clinical diagnosis success rate is around 80%. Thus, automated classification of these impairments in gait with machine learning techniques can serve as an assessment tool for identification of PD. The primary focus of the study is to investigate anthropometric parameter-based models for classification of non-PD and PD subjects together with their severity. The proposed work performs the computation of clinically relevant features using Vertical Ground Reaction Force (VGRF) data from a total of 165 individuals’ database consisting 93 Parkinson’s and 72 healthy controls. All extracted features are tested for significance and redundancy among other gait characteristics. The optimal combination of features is selected using Recursive Feature Elimination technique with 10-fold cross validation for classification. In this study, wide range of machine learning techniques from different domains are used and their performance is evaluated based on accuracy, specificity, recall, precision and F1-score. Both gender and age-gender specific models outperformed the generalized model for PD as well as PD severity assessment. The highest prediction accuracy reported for age-gender specific models is 98.50% (non-PD and PD) using Support Vector Machine (SVM) classifier and 97.76% (non-PD and PD with severity scales) using k-Nearest Neighbor (kNN) and SVM. This study demonstrates integration of gait data with machine learning techniques as a potential biomarker for assessment of PD severity.

Differentiating the learning styles of college students in different disciplines in a college English blended learning setting

Learning styles are critical to educational psychology, especially when investigating various contextual factors that interact with individual learning styles. Drawing upon Biglan's taxonomy of academic tribes, this study systematically analyzed the learning styles of 790 sophomores in a blended learning course with 46 specializations using a novel machine learning algorithm called the support vector machine (SVM). Moreover, an SVM-based recursive feature elimination (SVM-RFE) technique was integrated to identify the differential features among distinct disciplines. The findings of this study shed light on the optimal feature sets that collectively determined students' discipline-specific learning styles in a college blended learning setting.

Prediction of Land Suitability for Crop Cultivation Based on Soil and Environmental Characteristics Using Modified Recursive Feature Elimination Technique With Various Classifiers

Crop cultivation prediction is an integral part of agriculture and is primarily based on factors such as soil, environmental features like rainfall and temperature, and the quantum of fertilizer used, particularly nitrogen and phosphorus. These factors, however, vary from region to region: consequently, farmers are unable to cultivate similar crops in every region. This is where machine learning (ML) techniques step in to help find the most suitable crops for a particular region, thus assisting farmers a great deal in crop prediction. The feature selection (FS) facet of ML is a major component in the selection of key features for a particular region and keeps the crop prediction process constantly upgraded. This work proposes a novel FS approach called modified recursive feature elimination (MRFE) to select appropriate features from a data set for crop prediction. The proposed MRFE technique selects and ranks salient features using a ranking method. The experimental results show that the MRFE method selects the most accurate features, while the bagging technique helps accurately predict a suitable crop. The performance of proposed MRFE technique is evaluated by various metrics such as accuracy (ACC), precision, recall, specificity, F1 score, area under the curve, mean absolute error, and log loss. From the performance analysis, it is justified that the MRFE technique performs well with 95% ACC than other FS methods.