Interpretable Stroke Risk Prediction Using Machine Learning Algorithms

Zafeiropoulos, Nikolaos; Mavrogiorgou, Argyro; Kleftakis, Spyridon; Mavrogiorgos, Konstantinos; Kiourtis, Athanasios; Kyriazis, Dimosthenis

doi:10.1007/978-981-19-7663-6_61

Cited by 8 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is due to the limitation of LR does not perform well when the relationship between the input features and the target feature is highly non-linear, or when there are complex interactions between the features (Mordensky et al, 2023). As a part of future work, the study will consider dealing the combination of complexity features by implementing more analysis on ML such as Support Vector Machines (Mayes et al, 2023) and Naïve Bayes (Sawhney et al, 2023), k-Nearest Neighbor (Singh et al, 2022) and Random Forest (Zafeiropoulos et al, 2023) in solving the problem of e-ticketing system.…”

Section: Discussionmentioning

confidence: 99%

Solving Complexity Dataset in e-Ticketing using Machine Learning to Determine Optimum Feature

Mohd Jamaludin,

Majahar Ali,

Vun

et al. 2023

Mal. J. Fund. Appl. Sci.

View full text Add to dashboard Cite

e-ticketing is one of the common applications used in technical support in Information Technology (IT) and has been used worldwide in any field of company. The benefits of e-ticketing can reduce the human efforts, increase the sufficiency of system and provides the benefits and efficiency to the customers. Also, e-ticketing ticketing system can enhance worker safety, improving productivity, increasing project efficiency and cause the good impact on the performance of the business in terms of profitability. The main objective of this study is defining the model performance in each important feature by analysing the complex dataset by using logistic regression as a Machine Learning (ML) algorithm. In evaluation the performance of classifier, the dataset is injected to Python programming and split into 90% as training set and 10% for the testing set. From the analysis, the study found that only 3 out of 11 independent features in dataset that are relevant chosen to proceed the ML analysis. From the result, the accuracy for sct_short_description, sct_cmdb_ci, and sct_assignment_group is 41.65%, 48.77% and 96.49%, respectively. It showed that the accuracy’s result for the sct_assignment_group resulted that the model is very good accuracy and indicate that the model is well performing. Meanwhile, the value of F1-score is 96.11% in each feature. This result indicates that the model has a good balance of precision and recall in its binary classification predictions. Hence, the study considers the sct_assignment_group as a best features to proceed the analysis. The future study will consider dealing the combination of complexity features by implementing more analysis on ML such as Support Vector Machines and Naïve Bayes.

show abstract

Section: Discussionmentioning

confidence: 99%

Solving Complexity Dataset in e-Ticketing using Machine Learning to Determine Optimum Feature

Mohd Jamaludin,

Majahar Ali,

Vun

et al. 2023

Mal. J. Fund. Appl. Sci.

View full text Add to dashboard Cite

show abstract

“…The final performance is reported on the held-out test set in terms of Area Under the Receiver Operating Characteristic Curve (AUROC) [ 28 ], Area Under the Precision-Recall Curve (AUPRC) [ 29 ], and balanced accuracy with confidence intervals computed using bootstrapping with 1000 iterations. While there are a variety of metrics that can be reported for predictive models (e.g., precision, recall, specificity) [ 30 , 31 ], our choice of AUROC and AUPRC was driven by their ability to summarise the trade-off between commonly reported metrics at various thresholds. For instance, the AUROC metric quantifies the trade-off between specificity and sensitivity at various thresholds [ 32 ], while AUPRC summarizes the trade-off between precision and recall at various thresholds [ 29 ].…”

Section: Materials and Methodsmentioning

confidence: 99%

Synthesizing Electronic Health Records for Predictive Models in Low-Middle-Income Countries (LMICs)

2023

View full text Add to dashboard Cite

The spread of machine learning models, coupled with by the growing adoption of electronic health records (EHRs), has opened the door for developing clinical decision support systems. However, despite the great promise of machine learning for healthcare in low-middle-income countries (LMICs), many data-specific limitations, such as the small size and irregular sampling, hinder the progress in such applications. Recently, deep generative models have been proposed to generate realistic-looking synthetic data, including EHRs, by learning the underlying data distribution without compromising patient privacy. In this study, we first use a deep generative model to generate synthetic data based on a small dataset (364 patients) from a LMIC setting. Next, we use synthetic data to build models that predict the onset of hospital-acquired infections based on minimal information collected at patient ICU admission. The performance of the diagnostic model trained on the synthetic data outperformed models trained on the original and oversampled data using techniques such as SMOTE. We also experiment with varying the size of the synthetic data and observe the impact on the performance and interpretability of the models. Our results show the promise of using deep generative models in enabling healthcare data owners to develop and validate models that serve their needs and applications, despite limitations in dataset size.

show abstract

“…After XGBoost had been tested on the complete dataset, the idea of gain in XG was employed to determine each attribute's importance. XGBoost [55], which stan eXtreme Gradient Boosting, is a popular machine learning algorithm known for i ciency and effectiveness in handling large feature sets. It is commonly used for f selection tasks because of its ability to assess the importance of features in a datase…”

Section: Classification Of Relevant Features Using Xgboostmentioning

confidence: 99%

“…The recall is determined by dividing the number of TPs by the true positive examples in the dataset (TP + FP). The majority of the positive examples in the dataset, or all of the highly engaged students, are likely correctly identified by the model, according to a high recall score [55].…”

Section: Performance Matricesmentioning

confidence: 99%

Machine Learning-Based Adaptive Synthetic Sampling Technique for Intrusion Detection

2023

View full text Add to dashboard Cite

Traditional firewalls and data encryption techniques can no longer match the demands of current IoT network security due to the rising amount and variety of network threats. In order to manage IoT network risks, intrusion detection solutions have been advised. Even though machine learning (ML) helps the widely used intrusion detection techniques currently in use, these algorithms struggle with low detection rates and the requirement for extensive feature engineering. The deep learning model for IoT network intrusion detection is a method for traffic anomaly detection that is suggested by this study. To extract the sequence properties of data flow through a CNN, it combines an attention mechanism with a Long Short Term Memory (LSTM) network. This method uses adaptive synthetic sampling (ADASYN) to increase the size of minority-class samples. The proposed models demonstrated acceptable precision and recall for each class when used for binary-class classification, proving their stability and capacity to identify all classes correctly. The MLP classifier’s accuracy, precision, recall, and F1 value were 87%, 89%, 87%, and 89%, respectively, with an AUC score of 0.88. Overall, the proposed models performed well. The attack and all-class models exhibited good AUCs and macro metrics, the same as the proposed MLP classifier, which had an F1 score of 83% and an AUC score of 0.94. Additionally, it trained the MLP classifier and integrated the ADAM optimizer and category cross-entropy loss function for all-class classification. With an AUC value of 94%, it possessed 84% accuracy, 87% precision, 84% recall, and an 83% F1 score. A further indication of the hybrid model’s ability to combine the benefits of both models to improve overall performance was that it regularly outperformed the MLP model. This model’s accuracy and F1 score are better than those of earlier comparable algorithms, according to experimental results using the publicly accessible benchmark dataset for network intrusion detection (NSL–KDD).

show abstract

Interpretable Stroke Risk Prediction Using Machine Learning Algorithms

Cited by 8 publications

References 17 publications

Solving Complexity Dataset in e-Ticketing using Machine Learning to Determine Optimum Feature

Solving Complexity Dataset in e-Ticketing using Machine Learning to Determine Optimum Feature

Synthesizing Electronic Health Records for Predictive Models in Low-Middle-Income Countries (LMICs)

Machine Learning-Based Adaptive Synthetic Sampling Technique for Intrusion Detection

Contact Info

Product

Resources

About