State Key Laboratory for Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China.
Department of e-Systems, University of Bisha, Bisha 61922, Saudi Arabia.
Int J Environ Res Public Health. 2020 May 18;17(10):3510. doi: 10.3390/ijerph17103510.
Influenza epidemic is a serious threat to the entire world, which causes thousands of death every year and can be considered as a public health emergency that needs to be more addressed and investigated. Forecasting influenza incidences or confirmed cases is very important to do the necessary policies and plans for governments and health organizations. In this paper, we present an enhanced adaptive neuro-fuzzy inference system (ANFIS) to forecast the weekly confirmed influenza cases in China and the USA using official datasets. To overcome the limitations of the original ANFIS, we use two metaheuristics, called flower pollination algorithm (FPA) and sine cosine algorithm (SCA), to enhance the prediction of the ANFIS. The proposed FPASCA-ANFIS is evaluated using two datasets collected from the CDC and WHO websites. Furthermore, it was compared to some previous state-of-the-art approaches. Experimental results confirmed that the FPASCA-ANFIS outperformed the compared methods using variant measures, including RMSRE, MAPE, MAE, and R 2 .
流感疫情是对全球的严重威胁，每年导致数千人死亡，可被视为需要更多关注和研究的公共卫生紧急事件。预测流感发病率或确诊病例对于政府和卫生组织制定必要政策和计划非常重要。本文提出了一种增强型自适应神经模糊推理系统（ANFIS），使用官方数据集预测中国和美国的每周确诊流感病例。为了克服原始 ANFIS 的局限性，我们使用两种元启发式算法，即花授粉算法（FPA）和正弦余弦算法（SCA），来增强 ANFIS 的预测能力。所提出的 FPASCA-ANFIS 使用从 CDC 和世卫组织网站收集的两个数据集进行评估。此外，还与一些先前的最先进方法进行了比较。实验结果证实，FPASCA-ANFIS 在使用不同指标（包括均方根误差（RMSRE）、平均绝对误差（MAPE）、平均绝对误差（MAE）和 R 2 ）时，优于比较方法。
