Ranking of features for each algorithm.

Stroke analysis using game theory and machine learning techniques. The study investigates the use of the Shapley value in predictive ischemic brain stroke analysis. Initially, preference algorithms identify the most important features in various machine learning models, including logistic regression, K-nearest neighbor, decision tree, support vector machine (linear kernel), support vector machine ( RBF kernel), neural networks, etc. For each sample, the top 3, 4, and 5 features are evaluated and selected to evaluate their performance. The Shapley value method was used to rank the models using their best four features based on their predictive capabilities. As a result, better-performing models were found. Afterward, ensemble machine learning methods were used to find the most accurate predictions using the top 5 models ranked by shapely value. The research demonstrates an impressive accuracy of 92.39%, surpassing other proposed models’ performance. This study highlights the utility of combining game theory and machine learning in Ischemic stroke prediction and the potential of ensemble learning methods to increase predictive accuracy in ischemic stroke analysis.

本研究采用博弈论与机器学习技术开展脑卒中分析，重点探究夏普利值(Shapley value)在缺血性脑卒中预测分析中的应用。首先，偏好算法可在多种机器学习模型中识别出最为关键的特征，所涉模型包括逻辑回归(logistic regression)、K近邻(K-nearest neighbor)、决策树(decision tree)、线性核(linear kernel)支持向量机(Support Vector Machine, SVM)、径向基核(RBF kernel)支持向量机以及神经网络(neural networks)等。针对每个样本，研究将分别选取排名前3、前4及前5的特征，并对其对应的模型性能开展评估。随后，本研究基于各模型的4项最优特征，采用夏普利值法依据预测能力对模型进行排序，筛选出性能更优的模型。进一步地，本研究利用集成机器学习(ensemble machine learning)方法，结合经夏普利值排序得到的前5个最优模型，以获取精度更高的预测结果。实验结果显示，本方法取得了92.39%的出色准确率，优于其他已提出模型的性能表现。本研究证实了将博弈论与机器学习相结合用于缺血性脑卒中预测的有效性，同时也凸显了集成学习方法在提升缺血性脑卒中分析预测精度方面的应用潜力。