TY - JOUR JF - hums-jmis JO - JMIS VL - 3 IS - 2 PY - 2017 Y1 - 2017/10/01 TI - Relationship between socioeconomic status and burn patients survive using data mining TT - ارتباط‌سنجی وضعیت اجتماعی- اقتصادی بیماران سوختگی و حیات با استفاده از تکنیک‌های داده‌کاوی N2 - Aim: In all societies, burn is a health cause’s physiologic mental economic and health disorder both for the patient and its family challenge and more than any other trauma. Cases such as lifestyle, social, economic, and cultural levels of society can change the type of burns. Given that data mining is growing and new, it is also of great importance to medical data. Therefore, in this project, the intention is to collect data using actual data mining techniques and in view of the prevalence of burns in developing countries, a model for establishing a meaningful relationship between the economic and social status of individuals develop burns as a result of their treatment. Methods: The present study is analytic-descriptive based on retrospective nature. The population of the study consists of 553 patients all adult burn patients who were hospitalized in Ayatollah Taleghani hospital in Ahvaz for 3 years (1388-1390).data have been collected from check lists and questionnaire. The collected data were computed using spss22, IBM Modeler 14.2 and algorithm of C5.0, CHAID and CART. Results: The predicted model for the outcome of burn on patients conducted through selected algorithms in order of priority including factors of burn, occupation, capsule at home, place of incident, residential area, and rate of income, intent, gender and education. By comparing the accuracy of the algorithms, the highest accuracy is for the C & R algorithm (0.87), the most characteristic of the CHAID algorithm (0.98) and the most sensitivity was obtained for the C & R algorithm (0.50). Given the superiority of the C & R model, this model was recognized as the top model in terms of the accuracy and sensitivity of the models. Conclusion: Due to the average accuracy of suggested models, these models are both valid and attributable. The results of this study can be useful for the prediction of the impact of socioeconomic factors on burnout outcomes. SP - 18 EP - 25 AU - khedri, maasume AU - ghazi saeedi, marjan AU - Sheikh taheri, Abbas AU - zarei, javad AD - Department of Health Information Management, Faculty of Paramedicine, Tehran University of Medical Sciences, Tehran, Iran. KW - burn KW - socio-economic status KW - data mining UR - http://jmis.hums.ac.ir/article-1-104-en.html ER -