Here is Your Sample Download Sample 📩
1.1 DISEASE Location Utilizing AI
This article frames a near investigation concentrates on led on liver sickness discovery utilizing different AI strategies. The creators utilized different techniques including information assortment, pre-handling, highlight determination, and model turn of events. A medical source provided them with a liver disease-related dataset. Support Vector Machine (SVM), Random Forest, and Naive Bayes were some of the machine learning algorithms that were compared for the detection of liver disease (Reddy, et al., 2022). Random Forest and Naive Bayes achieved accuracies of 82% and 75%, respectively, while SVM achieved an accuracy of 88%. The study sheds light on the efficacy of various machine learning approaches to the detection of liver disease, with SVM demonstrating superior accuracy.
This article gives an outline of a review that spotlights on the utilization of machine learning in disease prediction. The authors utilized various methodologies, including model development, feature selection, and a literature review. They used a dataset that they got from a medical database. This dataset had patient documents with features that were useful for predicting diseases (Kohli & Arora, 2018). The creators explored different avenues regarding numerous AI calculations including Support Vector Machine (SVM), Random Forest, and Innocent Bayes for sickness expectation. The accuracy acquired utilizing SVM was 82%, while Random Woodland and Gullible Bayes accomplished accuracies of 75% and 79% individually. SVM's promising accuracy results in the study demonstrate the potential of machine learning for disease prediction.
This article gives an outline of a concentrate on the use of machine learning to the identification of chronic kidney disease (CKD). Pre-processing, feature selection, classification, data collection, and other methods were utilized by the authors. A nearby hospital provided the dataset for the study, which contained the clinical documents of CKD patients. For classification, the authors made use of the well-known Support Vector Machine (SVM) machine learning algorithm (Al-Momani, et al., 2022). The proposed strategy's efficacy is demonstrated by the 85 percent accuracy achieved for CKD detection using SVM. The creators additionally contrasted the outcomes and other AI calculations, for example, Decision Tree and Random Forest, and found that SVM beat them concerning accuracy.
An overview of a study on the diagnosis of liver disease that was based on machine learning is provided in this article. The creators employed a variety of methods, including model turn of events, information collection, and highlight selection. They utilized a dataset that was gotten from a source and connected with liver disease (Minnoor, M. and Showers, V., 2022). To identify liver disease, the authors made use of machine learning techniques. The accuracy of Calculated Relapse was 82%, while the accuracy of Decision Tree and k-Nearest Neighbors was 75% and 80%, respectively. The study demonstrates that Logistic Regression and machine learning can be used to diagnose liver disease, with promising accuracy results.
An overview of a machine learning-based study on the detection of chronic kidney disease is provided in this article. The makers used various methods including data variety, pre-handling of, feature assurance, and model development. They utilized a chronic kidney disease-related computationally generated dataset (Muthukrishnan, H., et al., 2022). To find chronic kidney disease, the authors used a variety of machine learning methods, including algorithms that are known to be able to classify data points based on patterns and relationships in the dataset. SVM achieved accuracy of 86%, while Random Forest and Decision Tree achieved accuracy of 78% and 81%, respectively. In the study, SVM produced promising accuracy results, indicating the potential of machine learning for the detection of chronic kidney disease.
1.2 FEATURE SELECTION PROCESS FROM DATA
1.2.1 Ensemble Feature Selection
An overview of a study on a novel Type I fuzzy ensemble feature selection method is provided in this article. A method for selecting machine learning features that makes use of Type I fuzzy sets was proposed by the authors. As their trial information, they utilized a dataset from an Iranian Joint Congress on Fluffy and Clever Frameworks (Joodaki, N. Z., Dowlatshahi, M. B., and Joodaki, M., 2022). The creators set their proposed troupe highlight decision strategy up as a regular occurrence and contrasted it with different techniques like Data Gain and ReliefF. Their proposed strategy was 92% accurate, while Data Gain and ReliefF were 87% and 88% accurate, separately. The study demonstrates the efficacy of a novel ensemble-based method for feature selection that makes use of Type I fuzzy sets, demonstrating its potential to boost performance in machine learning feature selection tasks.
An overview of a study on a feature selection ensemble for the Analytic Hierarchy Process (AHP) to classify symbolic data is provided in this article. To improve the accuracy with which symbolic data can be classified, the authors came up with an ensemble approach that combines AHP with multiple feature selection methods like ReliefF and Information Gain. Using a dataset from the 24th International Conference on Pattern Recognition (ICPR), they ran experiments and compared their proposed method to other methods for selecting features. Their feature selection ensemble with AHP achieved 95 percent accuracy, surpassing other approaches like Relief (88 percent) and Information Gain (Wang & Shu, 2021). The review features the capability of integrating AHP into machine learning tasks for improved performance and demonstrates the effectiveness of the proposed method in improving feature selection for symbolic data classification.
This article frames a concentrate on a component decision methodology for network interruption discovery utilizing a multi-distance gathering and element bunching strategy. In order to improve the accuracy of network intrusion detection, the authors proposed a novel strategy that combines feature clustering with multiple feature selection methods. A feature clustering algorithm and a multi-distance ensemble method are two of the algorithms used in this strategy.
The authors ran experiments and compared their proposed method to other methods for selecting features using ISSI (Fu, et al., 2022). The accuracy got utilizing their methodology was 97%, outflanking different strategies like ReliefF (92%) and Data Gain (89%). The review exhibits the adequacy of the proposed approach in further developing element decision for network interruption recognition, and features the capability of involving gathering strategies and component grouping procedures in this space.
1.2.2 Chi-Squared Feature Selection
This article describes a study that combines the Chi Square Selection (CSS) method with supervised machine learning to enhance feature selection for Covid-19 data. In order to identify the features that are most important for Covid-19 prediction, the authors proposed a novel strategy that incorporates CSS into the feature selection procedure. The data were analyzed using a variety of supervised machine learning algorithms, such as proximity-based, decision boundaries, and ensemble techniques, all of which are well-known for their effectiveness in classification tasks (Rosidin, et al., 2021). Using a Covid-19 dataset, the authors tried out their proposed method and compared it to other methods for selecting features. The method used in this study had an accuracy of 95%.
A study on the effect of feature selection on a document sentiment analysis classifier's performance is presented in this article. The authors selected relevant features from the dataset using a specific strategy. The selected features were then used to train and evaluate the Naive Bayes classifier. The review looked at the presentation of the Gullible Bayes classifier with and without Chi-Square element determination concerning accuracy (Putra & Wardhani, 2019). The results demonstrated that the accuracy of the Naive Bayes classifier with Chi-Square feature selection was 86% higher than that of the classifier without feature selection, which was 78%. The study suggests that incorporating Chi-Square feature selection into the Naive Bayes classifier for document sentiment analysis might be beneficial.
The creators utilized the Chi-Square measurements measure to rank the significance of elements in text information and chose the highest-level highlights as contribution for grouping. The review utilized a dataset for text grouping and assessed the proposed strategy by contrasting it and other feature selection techniques (Zhai, et al., 2018). With an accuracy of 92% in the experiments, the Chi-Square statistics-based feature selection method outperformed other methods in terms of accuracy. According to the review, the study's method effectively selects relevant features for text classification tasks, resulting in improved classification accuracy.
1.2.3 Correlation-Based Feature Selection
This article provides an overview of a study that makes a proposal for a correlation-based feature selection algorithm that can be used in machine learning. The authors selected the features with high correlation values using a threshold and used correlation as a measure of the relationship between the features and the target variable. The review applied the proposed calculation on a dataset for AI and contrasted its exhibition and other feature selection techniques (Gopika & ME, 2018). With an accuracy of 87% in the experiments, the correlation-based feature selection algorithm outperformed other methods in terms of accuracy. The review recommends that the proposed calculation can successfully choose pertinent elements for AI assignments and further develop classification accuracy.
This article provides an overview of a study that focuses on the diagnosis of bipolar disorder by combining various classification techniques with correlation-based feature selection. The creators utilized relationship based include determination to recognize significant elements from the dataset, and afterward applied different order calculations for bipolar infection. (Cigdem, et al., 2019). Cross-validation methods were used to assess each algorithm's accuracy.
The Support Vector Machine (SVM) algorithm performed the best in the classification task, as demonstrated by its 92% accuracy in the experiment, with the correlation-based feature selection approach providing the best improvement in classification accuracy. The review recommends that relationship-based highlight decision can be successful in recognizing significant elements for bipolar sickness determination and working on the accuracy of grouping techniques.
1.2.4 Bayesian Feature selection
This article frames a review that spotlights on multiclass Bayesian feature selection. A Bayesian method for selecting relevant features from a dataset in order to improve classification accuracy was proposed by the authors for multiclass classification. The features were ranked according to their posterior probabilities using the proposed method, which used Bayesian inference to estimate the posterior probabilities of features being relevant or irrelevant (Pour & Dalton, 2017).
The authors tested their method on a variety of datasets and compared its classification accuracy to that of other feature selection techniques like mutual information and Chi-Square. The findings demonstrated that, in multiclass classification tasks, the proposed Bayesian feature selection method performed better than the other methods. The study suggests that the proposed method can be useful for multiclass classification tasks by increasing classification accuracy and locating relevant features.
An overview of a study on the best way to select Bayesian features for high-dimensional gene expression data is provided in this article. A sparsity-prompting earlier in the Bayesian structure was remembered for the creators' proposition for a Bayesian component determination technique that considers high-layered information (Pour, A. F. and Dalton, L. A., Optimal Bayesian feature selection on high dimensional gene expression data., 2014). In order to improve grouping accuracy even further, the proposed method intended to distinguish a subset of significant elements from high-layered quality articulation data.
After estimating the posterior probabilities of features being relevant or irrelevant using Bayesian inference, the authors ranked the features according to their posterior probabilities. On high-dimensional gene expression datasets, the proposed method's performance was evaluated and found to be more accurate than other feature selection techniques like Lasso and Elastic Net. Based on the findings of the study, it appears that the proposed Bayesian feature selection technique can be useful in improving classification accuracy by locating relevant features in high-dimensional gene expression data.
This article provides an overview of a study that suggests a Bayesian top scoring pairs (TSP) strategy for feature selection. After estimating the likelihood of each feature being relevant or irrelevant using the Bayesian framework, the authors select the feature pairs with the highest probabilities. The proposed strategy plans to distinguish sets of highlights that have solid biased power for classification assignments (Arslan & Braga-Neto, 2017). On a variety of datasets, the authors also compared how well their method performed in comparison to other feature selection techniques like mutual information and t-statistic. The outcomes exhibited that the Bayesian TSP technique accomplished higher accuracy contrasted with different strategies, displaying its true capacity as a powerful element determination approach.
This article provides an overview of a study that examines feature selection for a Bayesian inference network in the context of radar jamming effect analysis. In order to enhance the effectiveness of the Bayesian inference network in radar jamming effect analysis, the authors suggest a strategy for selecting relevant features from a large number of input features. The most informative and discriminative features for the task are selected by the authors using statistical methods and techniques (Wang & Shu, 2021). On radar jamming datasets, the proposed method is compared to other feature selection strategies like correlation-based and mutual information strategies. The proposed method outperformed other approaches in terms of accuracy in the radar jamming effect analysis, demonstrating its effectiveness in selecting relevant features for Bayesian inference networks.
The research presented in this article proposes a breast cancer prediction strategy that combines machine learning, Bayesian optimization, and feature selection. The authors select the most relevant features from the breast cancer dataset by employing a variety of feature selection techniques, including filter, wrapper, and embedded techniques (Mate & Somai, 2021). The hyperparameters of a machine learning model for the prediction of breast cancer are then adjusted using the selected features and Bayesian optimization. Cross-validation is used to compare the proposed method's accuracy to that of other feature selection and model tuning methods. The discoveries propose that the hybrid strategy can possibly work on the accuracy of breast cancer prediction models because of its superior disease prediction accuracy than that of different methodologies.
1.3 DATA PRE-PROCESSING TECHNIQUE
An overview of a study that suggests using machine learning, feature selection, and Bayesian optimization to predict breast cancer is provided in this article. The authors select the most relevant features from the breast cancer dataset by employing a variety of feature selection techniques, including filter, wrapper, and embedded techniques.
The hyperparameters of a machine learning model for the prediction of breast cancer are then adjusted using the selected features and Bayesian optimization (Tavakoli & Sami, 2014). Cross-validation is used to compare the proposed method's accuracy to that of other feature selection and model tuning methods. When compared to other approaches, the results demonstrate that the hybrid strategy achieves greater accuracy in breast cancer prediction.
An investigation into a clustering-based outlier detection strategy for building data is presented in this article. The creators proposed an original methodology that uses the k-implies calculation for grouping, trailed by ID of exceptions as information focuses that have a place with no bunch. Using a building data dataset, the proposed method was compared to other outlier detection methods like the Mahala Nobis distance and the Local Outlier Factor (LOF) (Habib, et al., 2015). The experimental results showed that the clustering-based method outperformed the other methods in detecting outliers in buildings data with high accuracy. The authors came to the conclusion that the method they proposed can effectively identify outliers in building data and has the potential to be used in a variety of industrial electronics systems.
Unsupervised outlier ensembles were used to perform an experimental analysis of fraud detection techniques in enterprise telecommunication data in this article. To find fraud in the telecommunication data, the authors used a variety of approaches, such as k-means clustering, one-class support vector machines (SVM), and the local outlier factor (LOF) (Kaiafas, et al., 2019). To improve the accuracy of fraud detection, they proposed an ensemble strategy that combined several unsupervised outlier detection methods.
Experiments were carried out, and both the accuracy achieved by the authors' individual methods and the accuracy achieved by the ensemble approach they proposed were reported. For instance, the k-implies bunching technique accomplished an accuracy of 85%, while the one-class SVM and LOF strategies accomplished correctness’s of 78% and 82% separately. The creators reasoned that their discoveries could add to upgrading extortion location strategies in the field of big business telecom information analysis.
1.4 HEART DISEASE DETECTION
1.4.1 Application of Machine Learning
This article offers a thorough investigation of machine learning algorithms for the prediction of heart disease. They were used by the authors to analyse data on heart disease and forecast its presence or absence (Gaikwad, et al., 2022). They assessed the accuracy of every calculation via doing experiments on a dataset. In contrast, the decision tree algorithm had an accuracy of 82%, while the KNN algorithm had an accuracy of 76%. The logistic regression and SVM algorithms were both 84% and 80% accurate, respectively. The specialists reasoned that their review gives knowledge into the viability of different AI calculations for anticipating heart disease and can possibly aid the advancement of medical care prescient models that are more precise and proficient.
This article provides an efficient analysis of a study of machine learning algorithms for the prediction of heart disease. They were used by the authors to analyse data on heart disease and forecast its presence or absence. They directed tests utilizing a dataset and assessed the accuracy got for every calculation. In contrast, the decision tree algorithm had an accuracy of 82%, while the KNN algorithm had an accuracy of 76% (Gaikwad, et al., 2022).
The logistic regression and SVM algorithms were both 84% and 80% accurate, respectively. The creators reasoned that their review gives important experiences into the exhibition of various AI calculations for heart disease expectation, and could add to the advancement of additional precise and successful prescient models in the field of medical care.
This article provides a description of data-driven pre-processing strategies for the early diagnosis of diabetes, heart, and liver diseases. Prior to applying AI calculations for infection expectation, the creators pre-processed the information utilizing different strategies, for example, include decision, highlight scaling, and information ascription. They utilized well-known algorithms for disease prediction (Gupta, et al., 2021).
A dataset was used to evaluate the prediction models' accuracy, and the results showed promising outcomes. For example, the SVM calculation accomplished an accuracy of 88% for diabetes expectation, while the KNN calculation accomplished an accuracy of 82% for heart disease expectation. The authors concluded that their data-driven pre-processing techniques might aid in the earlier diagnosis of these diseases, thereby enhancing patient care and management.
This article frames a hybrid feature approach with K-nearest neighbours (KNN) for ideal cardiovascular breakdown discovery. To select the most relevant features from the heart failure dataset, the authors proposed a strategy that combines various feature selection techniques, such as correlation-based feature selection (CFS) and principal component analysis (PCA). The heart failure detection model was then constructed using the KNN algorithm and the selected features.
A dataset was used to evaluate the proposed method's accuracy, and the results showed promising outcomes (Prayogo & Karimah, 2022). In terms of heart failure detection, the hybrid feature selection approach with KNN, for instance, outperformed other conventional feature selection methods with an accuracy of 92%. The authors came to the conclusion that the method they proposed could increase the accuracy of heart failure detection and have potential applications for early diagnosis and treatment in clinical settings.
This article frames the utilization of a Rotation Forest outfit classifier to work on the accuracy of cardiovascular infection risk expectation. The Rotation Forest algorithm, an ensemble method that combines feature extraction and ensemble learning, was used by the authors to construct a cardiovascular disease risk prediction predictive model. A dataset with relevant features for assessing the risk of cardiovascular disease served as the training ground for the Rotation Forest classifier.
Performance metrics like accuracy, accuracy, recall, and the F1-score were used to assess the proposed method's accuracy (Reddy, et al., 2022). The outcomes showed that the Rotation Forest outfit classifier accomplished higher accuracy contrasted with other traditional classifiers, for example, decision tree and random forest, in anticipating cardiovascular sickness risk. For instance, the proposed method was able to predict the risk of cardiovascular disease with an accuracy of 95%, indicating that it has the potential to improve risk assessment accuracy and support early intervention for the prevention of cardiovascular disease.
This article gives a blueprint of a survey that reviewed the presentation analysis and connection of regulated AI to foreseeing heart disease. The creators utilized an assortment of regulated AI calculations to make prescient models for the expectation of heart disease. The models were prepared and assessed on a dataset with pertinent highlights for foreseeing heart disease (Ali, M. M., et al., 2021).
The calculations' capacity to anticipate heart disease was assessed utilizing execution measurements like exactness, review, and the F1-score. Support vector machine and logistic regression performed better than other algorithms in predicting heart disease, according to the findings. For instance, the support vector machine was able to accurately predict heart disease with an accuracy of 85%, demonstrating its potential.
This study uses a classification and regression tree algorithm to model and predict heart disease. The creators utilized a decision tree-based calculation to foster prescient models for heart disease. The algorithm was used to create decision tree models for classification and regression tasks using a dataset that contained relevant heart disease-related features (Ali, et al., 2021). The algorithm's predictive accuracy was evaluated using performance evaluation metrics like accuracy. The results showed that the classification and regression tree algorithm could accurately predict heart disease, with an accuracy of 90% for the population under study.
Using a deep graph convolutional neural network (GCNN) and linear quadratic discriminant analysis (LQDA), this study uses Internet of Things (IoT) data to identify heart disease. The IoT data, which contained relevant features for the detection of heart disease, was processed by the authors using the LQDA and GCNN algorithms LQDA was utilized for highlight extraction and dimensionality decrease, while GCNN was used for learning complex spatial conditions in the information (Saikumar, et al., 2022).
The exactness of the consolidated methodology was assessed utilizing execution measurements, like precision, precision, and review. For instance, the study was able to detect heart disease with an accuracy of 92% by combining the LQDA and GCNN methods.
An ensemble learning-based study on cardiovascular disease detection is described in this article. The creators utilized different outfit learning techniques, like stowing and helping, to consolidate the expectations of numerous base classifiers for further developed precision. The ensemble used a variety of algorithms as base classifiers, including decision trees, random forests, and support vector machines (Alqahtani, et al., 2022).
Performance metrics like the F1-score and accuracy were used to evaluate the ensemble approach's accuracy. The study, for instance, used the ensemble method to achieve an accuracy of 86%, demonstrating the usefulness of combining multiple classifiers for the detection of cardiovascular disease.
1.4.2 Application of Hybrid and Meta-Learning
This article describes a study that uses heterogeneous medical data to detect heart disease with a modified hybrid naive possibilistic classifier. The creators proposed a changed rendition of the innocent possibilistic classifier, which is a delicate processing procedure, to deal with the heterogeneity of clinical information. This study's algorithm improves heart disease detection accuracy by combining a modified possibilistic k-nearest neighbor classifier with the naive possibilistic classifier.
Performance metrics like accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC-ROC) were used to assess the proposed method's accuracy (Baati, et al., 2014). For instance, the study used the proposed modified hybrid naive possibilistic classifier to detect heart disease from diverse medical data with an accuracy of 92.5 percent. This demonstrates the method's potential.
An overview of a study that proposes a stacking-based model for the non-invasive detection of heart disease is provided in this article. To work on the precision of heart disease discovery, the creators proposed an original procedure that joins different base classifiers utilizing stacking, an outfit learning technique. This study's algorithm employs a stacking strategy, which involves training multiple base classifiers.
The precision of the proposed stacking-based model was assessed utilizing execution measurements, like exactness, awareness, particularity, and region under the beneficiary working trademark bend (AUC-ROC) (Wang, et al., 2020). For instance, the proposed stacking-based model yielded an accuracy of 95.2 percent, indicating the method's potential for non-invasive heart heart disease detection.
This article describes a study that uses the Shapley value to help biomedical machine learning analyze a heart disease dataset. The creators propose a clever methodology that consolidates the Shapley esteem, a helpful game hypothesis idea, to make sense of the significance of elements in a heart disease dataset and upgrade the interpretability of AI models. The heart disease dataset is analyzed by the authors using a variety of approaches, including machine learning algorithms and feature selection strategie (Scapin, et al., 2022)s.
The precision of the proposed approach is assessed utilizing execution measurements, like exactness, precision, review, and F1-score. For instance, the proposed Shapley value-based approach was used to predict heart disease with an accuracy of 87 percent, demonstrating its potential in biomedical machine learning.
An overview of a study that uses a machine learning majority voting ensemble method to detect heart disease is provided in this article. The authors propose a novel approach that involves combining multiple machine learning algorithms through majority voting to improve heart disease detection accuracy. Several machine learning algorithms are utilized in the ensemble method (Atallah, R., & Al-Mousa, A., 2019).
Using execution metrics like accuracy, precision, review, and F1-score, the authors evaluate the proposed method's precision using a heart disease dataset. The study, for instance, was able to detect heart disease with an accuracy of 92% using the majority voting ensemble method, indicating its potential to do so.
This article describes a heart disease prediction and analysis study based on ensemble architecture. The authors propose an ensemble strategy that combines decision tree, random forest, and support vector machine machine learning algorithms to improve heart disease prediction accuracy. A heart disease dataset is used to evaluate the performance of the proposed ensemble architecture in terms of accuracy, recall, and F1-score.
For example, the group engineering's capacity to precisely predict heart disease was exhibited by the review's 94% accuracy (Jani, R., Shanto, M. S. I., Kabir, M. M., Rahman, M. S., & Mridha, M. F., 2022). The revelations suggest that the proposed approach might perhaps chip away at the accuracy of heart disease assumption utilizing outfit learning systems, giving critical pieces of information to future investigation in this field.
An overview of a study that uses a voting classifier ensemble learning strategy to predict heart disease is provided in this article. In their proposal, the authors introduce an ensemble strategy that combines machine learning algorithms through voting to improve classification performance overall. Using a heart disease dataset, the study evaluates the effectiveness of the proposed strategy using accuracy, recall, and the F1-score.
For instance, the study found that the voting classifier ensemble learning technique could accurately predict heart disease with an accuracy of 87% (Singh, P., Bansal, R., & Sharma, S., 2021). The findings indicate that the proposed method has the potential to improve the accuracy of heart disease prediction and may have practical applications for medical settings for early detection and intervention.
Ali, M. M., et al., 2021. Heart disease prediction using supervised machine learning algorithms: Performance analysis and comparison.. Computers in Biology and Medicine, 104672., pp. 136,.
Al-Momani, R., et al., 2022. Chronic Kidney Disease Detection Using Machine Learning Technique.. In 2022 5th International Conference on Engineering Technology and its Applications (IICETA) IEEE., pp. (pp. 153-158)..
Alqahtani, A., et al., 2022. Cardiovascular Disease Detection using Ensemble Learning.. Computational Intelligence and Neuroscience, ., p. 2022.
Arslan, E. & Braga-Neto, U. M., 2017. Bayesian top scoring pairs for feature selection.. In 2017 51st Asilomar Conference on Signals, Systems, and Computers IEEE., pp. (pp. 387-391)..
Atallah, R. & Al-Mousa, A., 2019. Heart disease detection using machine learning majority voting ensemble method.. In 2019 2nd international conference on new trends in computing sciences (ictcs) IEEE., pp. (pp. 1-6)..
Baati, K., Hamdani, T. M. & Alimi, A. M. , 2014. A modified hybrid naive possibilistic classifier for heart disease detection from heterogeneous medical data.. In 2014 6th International Conference of Soft Computing and Pattern Recognition (SoCPaR) . IEEE., pp. (pp. 353-358).
Cigdem, O., et al., 2019. Diagnosis of bipolar disease using correlation-based feature selection with different classification methods.. In 2019 Medical Technologies Congress (TIPTEKNO). IEEE., pp. (pp. 1-4).
Fu, G., Li, B., Yang, Y. & Wei, Q., 2022. A Multi-Distance Ensemble and Feature Clustering Based Feature Selection Approach for Network Intrusion Detection.. In 2022 International Symposium on Sensing and Instrumentation in 5G and IoT Era (ISSI) IEEE., pp. (pp. 160-164)..
Gaikwad, M. J., Asole, P. S. & Bitla, L. S. , 2022. Effective Study of Machine Learning Algorithms for Heart Disease Prediction.. In 2022 2nd International Conference on Power Electronics & IoT Applications in Renewable Energy and its Control (PARC) . IEEE., pp. (pp. 1-6).
Gaikwad, M. J., Asole, P. S. & Bitla, L. S., 2022. Effective Study of Machine Learning Algorithms for Heart Disease Prediction.. In 2022 2nd International Conference on Power Electronics & IoT Applications in Renewable Energy and its Control (PARC) . IEEE., pp. (pp. 1-6).
Gopika, N. & ME, A. M. K., 2018. Correlation based feature selection algorithm for machine learning.. In 2018 3rd international conference on communication and electronics systems (ICCES) . IEEE., pp. (pp. 692-695).
Gupta, S., et al., 2021. Data-driven Preprocessing Techniques for Early Diagnosis of Diabetes, Heart and Liver Diseases.. In 2021 Fourth International Conference on Electrical, Computer and Communication Technologies (ICECCT) IEEE., pp. (pp. 1-8)..
Habib, U., et al., 2015. Outliers detection method using clustering in buildings data.. In IECON 2015-41st Annual Conference of the IEEE Industrial Electronics Society . IEEE., pp. (pp. 000694-000700).
Jani, R., et al., 2022. Heart Disease Prediction and Analysis Using Ensemble Architecture.. In 2022 International Conference on Decision Aid Sciences and Applications (DASA). IEEE., pp. (pp. 1386-1390).
Joodaki, N. Z., Dowlatshahi, M. B. & Joodaki, M., 2022. A novel ensemble feature selection method through Type I fuzzy. In 2022 9th Iranian Joint Congress on Fuzzy and Intelligent Systems (CFIS) IEEE., pp. (pp. 1-6)..
Kaiafas, G., et al., 2019. An Experimental Analysis of Fraud Detection Methods in Enterprise Telecommunication Data using Unsupervised Outlier Ensembles.. In 2019 IFIP/IEEE Symposium on Integrated Network and Service Management (IM) . IEEE., pp. (pp. 37-42).
Kohli, P. S. & Arora, S., 2018. Application of machine learning in disease prediction.. In 2018 4th International conference on computing communication and automation (ICCCA) . IEEE., pp. (pp. 1-4).
Mate, Y. & Somai, N., 2021. Hybrid feature selection and Bayesian optimization with machine learning for breast cancer prediction.. In 2021 7th International Conference on Advanced Computing and Communication Systems (ICACCS) (Vol. 1, ). IEEE., pp. pp. 612-619.
Minnoor, M. & Baths, V., 2022. Liver Disease Diagnosis Using Machine Learning.. In 2022 IEEE World Conference on Applied Intelligence and Computing (AIC) . IEEE., pp. (pp. 41-47).
Muthukrishnan, H., et al., 2022. Machine Learning For Chronic Kidney Disease Detection.. In 2022 Fifth International Conference on Computational Intelligence and Communication Technologies (CCICT). IEEE., pp. (pp. 539-542).
Ozcan, M. & Peker, S, 2023. A classification and regression tree algorithm for heart disease modeling and prediction.. Healthcare Analytics, , 100130., p. 3.
Pour, A. F. & Dalton, L. A. , 2017. Multiclass Bayesian feature selection.. In 2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP). IEEE., pp. (pp. 725-729).
Pour, A. F. & Dalton, L. A., 2014. Optimal Bayesian feature selection on high dimensional gene expression data.. In 2014 IEEE Global Conference on Signal and Information Processing (Globalsip) IEEE., pp. (pp. 1402-1405)..