{"title":"使用机器学习模拟环境对公众健康的影响:哮喘案例研究","authors":"Lakmini Wijesekara, L. Liyanage","doi":"10.1109/CITISIA50690.2020.9397488","DOIUrl":null,"url":null,"abstract":"Environmental conditions such as weather and pollution have direct links with public health. It is estimated that the global burden of disease attributed to environmental factors is 24%. A plethora of research has been carried out to investigate the links between the environment and public health. Most of them are clinical or experimental studies. In addition to the investigations of causal effects, it is always useful to study associations of weather and pollution with diseases to manage and mitigate the burden of diseases as well as other environmental issues holistically. Environmental conditions could be used to provide an alarm of a future episode of a disease such as asthma so that risky individuals can take precautions to minimize the risk. This study involves a case study of asthma which applies several machine learning techniques to build a classification model predicting the risk of getting future episodes of asthma based on weather and pollution conditions. Support Vector Machine, Artificial Neural Network, Decision Tree and Random Forest models were considered for the classification. Random forest model produced the best performance compared to other models with an accuracy of 77%. Decision tree model exhibits the highest sensitivity of 70%. Even though ANN gives the lowest accuracy of 59%, its learning curve shows a good fit.","PeriodicalId":145272,"journal":{"name":"2020 5th International Conference on Innovative Technologies in Intelligent Systems and Industrial Applications (CITISIA)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Modelling Environmental Impact on Public Health using Machine Learning: Case Study on Asthma\",\"authors\":\"Lakmini Wijesekara, L. Liyanage\",\"doi\":\"10.1109/CITISIA50690.2020.9397488\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Environmental conditions such as weather and pollution have direct links with public health. It is estimated that the global burden of disease attributed to environmental factors is 24%. A plethora of research has been carried out to investigate the links between the environment and public health. Most of them are clinical or experimental studies. In addition to the investigations of causal effects, it is always useful to study associations of weather and pollution with diseases to manage and mitigate the burden of diseases as well as other environmental issues holistically. Environmental conditions could be used to provide an alarm of a future episode of a disease such as asthma so that risky individuals can take precautions to minimize the risk. This study involves a case study of asthma which applies several machine learning techniques to build a classification model predicting the risk of getting future episodes of asthma based on weather and pollution conditions. Support Vector Machine, Artificial Neural Network, Decision Tree and Random Forest models were considered for the classification. Random forest model produced the best performance compared to other models with an accuracy of 77%. Decision tree model exhibits the highest sensitivity of 70%. Even though ANN gives the lowest accuracy of 59%, its learning curve shows a good fit.\",\"PeriodicalId\":145272,\"journal\":{\"name\":\"2020 5th International Conference on Innovative Technologies in Intelligent Systems and Industrial Applications (CITISIA)\",\"volume\":\"17 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-11-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 5th International Conference on Innovative Technologies in Intelligent Systems and Industrial Applications (CITISIA)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CITISIA50690.2020.9397488\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 5th International Conference on Innovative Technologies in Intelligent Systems and Industrial Applications (CITISIA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CITISIA50690.2020.9397488","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Modelling Environmental Impact on Public Health using Machine Learning: Case Study on Asthma
Environmental conditions such as weather and pollution have direct links with public health. It is estimated that the global burden of disease attributed to environmental factors is 24%. A plethora of research has been carried out to investigate the links between the environment and public health. Most of them are clinical or experimental studies. In addition to the investigations of causal effects, it is always useful to study associations of weather and pollution with diseases to manage and mitigate the burden of diseases as well as other environmental issues holistically. Environmental conditions could be used to provide an alarm of a future episode of a disease such as asthma so that risky individuals can take precautions to minimize the risk. This study involves a case study of asthma which applies several machine learning techniques to build a classification model predicting the risk of getting future episodes of asthma based on weather and pollution conditions. Support Vector Machine, Artificial Neural Network, Decision Tree and Random Forest models were considered for the classification. Random forest model produced the best performance compared to other models with an accuracy of 77%. Decision tree model exhibits the highest sensitivity of 70%. Even though ANN gives the lowest accuracy of 59%, its learning curve shows a good fit.