...
首页> 外文期刊>Journal of Theoretical and Applied Information Technology >FORECASTING THE NUMBER OF DENGUE FEVER CASES IN MALANG REGENCY INDONESIA USING FUZZY INFERENCE SYSTEM MODELS
【24h】

FORECASTING THE NUMBER OF DENGUE FEVER CASES IN MALANG REGENCY INDONESIA USING FUZZY INFERENCE SYSTEM MODELS

机译:应用模糊推理系统模型预测玛琅再生印度尼西亚登革热病例数。

获取原文
           

摘要

Dengue fever is a communicable disease that has been a big concern in Indonesia. This disease has spread out across Indonesia, including Malang Regency. Local Government and Public Health Service in Malang Regency has made various efforts including prevention and socialization, however the number of casualties caused by dengue fever are still high. Forecasting the number of dengue fever cases is very important for the local Public Health Service. It can help policy planning of disease prevention and patient care in the future. Delays in preventive measures, increasing casualties, lack of treatment facilities are the problems that can be avoided through better policy planning. In this research, Fuzzy Inference System (FIS) is used to predict number of dengue fever cases in Malang. FIS tends to have small error values and high accuracy due to detailed attention to all variables. Fuzzy Inference System does not require a lot of data and a long periods of time. The model is constructed by grouping the number of monthly dengue fever cases from the previous years based on geographical location. Population density is added as external variables of the model. The data is divided into training set, testing set, and validating set with the ratio of 70:20:10. This research shows that forecasting model based on FIS shows a good results in forecasting with MAPE 6% in lowlands, 12% in mediumlands, and 14% in highlands.
机译:登革热是一种传染性疾病,在印尼引起很大关注。这种疾病已经蔓延到包括玛琅摄政在内的整个印度尼西亚。玛琅摄政区的地方政府和公共卫生服务部门已经做出了各种努力,包括预防和社会化,但是登革热造成的伤亡人数仍然很高。预测登革热病例数对于当地公共卫生服务非常重要。它可以帮助将来进行疾病预防和患者护理的政策规划。预防措施的拖延,人员伤亡的增加,治疗设施的缺乏是可以通过更好的政策计划来避免的问题。在这项研究中,模糊推理系统(FIS)用于预测玛琅的登革热病例数。由于对所有变量的关注,FIS倾向于具有较小的误差值和较高的精度。模糊推理系统不需要大量数据,也不需要很长时间。该模型是根据地理位置将过去几年中每月的登革热病例数进行分组而构建的。人口密度被添加为模型的外部变量。数据按70:20:10的比例分为训练集,测试集和验证集。这项研究表明,基于FIS的预测模型在低海拔地区的MAPE预测为6%,中高地为12%,高地为14%的情况下具有良好的预测效果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号