Ensemble Learning for Data-Driven Diagnosis of Polycystic Ovary Syndrome

E-GCS: Detection of COVID-19 through classification by attention bottleneck residual network

“… Methods Recall (in %) Precision (in %) Validation accuracy (in %) Vanilla gray 58 68 67.80 Vanilla RGB 61 68 69 Hybrid CNN VGG 62 68 69.50 VDSNet 63 69 73 Modified CapsNet 48 61 63.80 Basic CapsNet 51 64 60.50 Proposed 96.64 95 96.02

Fig. 12 Analysis in terms of performance metrics ( Bharati et al, 2020b ).

…”

“…In addition, two forms of optimised NASNet (Neural Architecture Search Network) have been suggested to diagnose COVID-19. Outcomes have explored that, suggested NASNet-mobile exposed 82.42% as accuracy, while, NASNet-large has exposed 81.06% as accuracy ( Bharati et al, 2020b ). Further, an optimised Inception ResNetV2 has been suggested to detected COVID-19.…”

Exploring the dominant features and data-driven detection of polycystic ovary syndrome through modified stacking ensemble machine learning technique

“…Most of the previous studies in this area were based on traditional ML classifiers. However, recently a few researchers have focused on applying ensemble techniques in PCOS detection, but their exploration techniques are based on typical bagging, boosting or voting type of ensemble models [29] , [45] . To the best of our knowledge, the proposed technique based on stacking ensemble classification approach where both traditional as well as boosting or bagging ensemble models are aggregated to provide a stronger prediction is a unique solution in this domain.…”

“…A few studies have been found where ensemble techniques had been employed in PCOS identification, for example Gupta et al [44] applied four types of Boosting ensemble techniques (Adaptive Boost, Gradient Boost, XGBoost and CatBoost) without applying any feature engineering techniques to classify PCOS. Again, Bharati et al [45] applied hard and soft voting ensemble classifier employing ExtraTree, Random Forest, Gaussian Naive Bayes, LightGBM and eXtreme Gradien Boosting models with reduced set of features selected via recursive feature elimination and univariate feature selection techniques. However, From the prior studies, it can be demonstrated that even though many academics from around the world have suggested contributions where various machine learning strategies have been used to diagnose PCOS; seldom has a researcher looked into the viability and effectiveness of using several ensemble machine learning approaches (bagging, boosting, and stacking) in this circumstance.…”

Federated learning: Applications, challenges and future directions

“…It is not uncommon for medical FL apps to attempt classification. Classification algorithms in machine learning [70] learn how to categorize or annotate a given collection of occurrences with labels or classes. For example, the classification tasks of COVID-19 detections [71][72][73][74][75][76][77][78][79][80][81][82][83][84], cancer diagnoses [85][86][87][88][89][90][91][92][93][94] and autism spectrum disorder (ASD) [84,[95][96][97] are considered in a FL setting in healthcare.…”