【摘要】 提出一种群优化拟合方法,通过测试函数和优化模型,分析验证了其具有较好的优化能力。采用群优化拟合方法计算了关键特征集合,并结合临床数据提出了一种基于群优化拟合及临床数据的癌症lncRNA预测方法。该方法在关键特征集合的基础上采用判别分析完成预测,预测过程中采用马氏统计距离的最小原则。实验结果表明,该方法获得了较好的收敛性能,在精确度、召回率和F1-Score三个指标上都达到了较好的预测结果。更多还原

【Abstract】 In this paper,a method of swarm optimization fitting was proposed,which was proved to have good optimization ability by test function and optimization model.The pivotal feature set was calculated by the method of swarm optimization fitting,and a prediction method of cancer lncRNA based on swarm optimization fitting and clinical data was proposed in combination with clinical data.The method used discriminant analysis to complete the prediction based on pivotal feature set,and the Mahalanobis statistical distance principle was adopted in the prediction process.Experimental results show that this method achieved good convergence performance and good prediction results in accuracy,recall rate and F1-Score.更多还原