{"title":"直接高阶模糊规则分类系统在死亡率预测中的应用","authors":"A. D. Torshizi, L. Petzold, M. Cohen","doi":"10.1109/BIBM.2015.7359795","DOIUrl":null,"url":null,"abstract":"Trauma is one of the leading causes of death in the U.S. and is ranked third among death causes across all age groups. This paper presents a novel fuzzy rule-based classification approach based on the concept of General Type-2 Fuzzy sets to predict mortality for trauma patients. In this approach each rule in the rule-base has an IF and a THEN part and parameters of the IF part (antecedents) are automatically extracted using powerful general type-2 fuzzy clustering algorithms which enables the model to deal with noisy and/or missing data. To verify efficacy of the proposed model, it has been implemented on several publicly available datasets. Finally, it is used to predict mortality among patients having traumatic injuries based on a large clinical dataset. Accuracy results demonstrate superior capabilities of the proposed approach compared to crisp and fuzzy classification methods in the literature.","PeriodicalId":186217,"journal":{"name":"2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"175 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Direct higher order fuzzy rule-based classification system: Application in mortality prediction\",\"authors\":\"A. D. Torshizi, L. Petzold, M. Cohen\",\"doi\":\"10.1109/BIBM.2015.7359795\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Trauma is one of the leading causes of death in the U.S. and is ranked third among death causes across all age groups. This paper presents a novel fuzzy rule-based classification approach based on the concept of General Type-2 Fuzzy sets to predict mortality for trauma patients. In this approach each rule in the rule-base has an IF and a THEN part and parameters of the IF part (antecedents) are automatically extracted using powerful general type-2 fuzzy clustering algorithms which enables the model to deal with noisy and/or missing data. To verify efficacy of the proposed model, it has been implemented on several publicly available datasets. Finally, it is used to predict mortality among patients having traumatic injuries based on a large clinical dataset. Accuracy results demonstrate superior capabilities of the proposed approach compared to crisp and fuzzy classification methods in the literature.\",\"PeriodicalId\":186217,\"journal\":{\"name\":\"2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)\",\"volume\":\"175 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-11-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/BIBM.2015.7359795\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/BIBM.2015.7359795","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Direct higher order fuzzy rule-based classification system: Application in mortality prediction
Trauma is one of the leading causes of death in the U.S. and is ranked third among death causes across all age groups. This paper presents a novel fuzzy rule-based classification approach based on the concept of General Type-2 Fuzzy sets to predict mortality for trauma patients. In this approach each rule in the rule-base has an IF and a THEN part and parameters of the IF part (antecedents) are automatically extracted using powerful general type-2 fuzzy clustering algorithms which enables the model to deal with noisy and/or missing data. To verify efficacy of the proposed model, it has been implemented on several publicly available datasets. Finally, it is used to predict mortality among patients having traumatic injuries based on a large clinical dataset. Accuracy results demonstrate superior capabilities of the proposed approach compared to crisp and fuzzy classification methods in the literature.