مقایسه مدل لجستیک و درخت طبقه بندی : یک برنامه کاربردی برای داده های افسردگی پس از زایمان
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|24721||2007||8 صفحه PDF||سفارش دهید||4398 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 32, Issue 4, May 2007, Pages 987–994
In this study, it is aimed that comparing logistic regression model with classification tree method in determining social-demographic risk factors which have effected depression status of 1447 women in separate postpartum periods. In determination of risk factors, data obtained from prevalence study of postpartum depression were used. Cut-off value of postpartum depression scores that calculated was taken as 13. Social and demographic risk factors were brought up by helping of the classification tree and logistic regression model. According to optimal classification tree total of six risk factors were determined, but in logistic regression model 3 of their effect were found significantly. In addition, during the relations among risk factors in tree structure were being evaluated, in logistic regression model corrected main effects belong to risk factors were calculated. In spite of, classification success of maximal tree was found better than both optimal tree and logistic regression model, it is seen that using this tree structure in practice is very difficult. But we say that the logistic regression model and optimal tree had the lower sensitivity, possibly due to the fact that numbers of the individuals in both two groups were not equal and clinical risk factors were not considered in this study. Classification tree method gives more information with detail on diagnosis by evaluating a lot of risk factors together than logistic regression model. But making correct selection through constructed tree structures is very important to increase the success of results and to reach information which can provide appropriate explanations.
Classification methods are commonly used in medicine particularly with the purpose of diagnosing (Harper et al., 2003). Usability of these methods increases parallel with developments in statistical packet programs. These methods usually evaluate more than one variable together and are examined in multivariate analyses group. If dependent variable consists of two (binary) or more (multinomial) categories, taking more than one risk factor or predictor variables together into the model with the purpose of estimating the values of dependent variable or correct classifying that will be increased the success in classification. Classification models are being used commonly with this purpose in discriminant analysis, logistic regression analysis, cluster analysis and neural network (Breiman et al., 1984, Cappelli et al., 1998 and Hosmer and Lemeshow, 1989). Logistic regression and Classification Trees (CT) are the models being used for estimating class membership of categorical dependent variable without getting any assumption on independent variables (Breiman et al., 1984, Buntine, 1992, Cappelli et al., 2002, Hosmer and Lemeshow, 1989, Kerby, 2003, Olaru and Wehenkel, 2003, Siciliano and Mola, 2000 and Terin et al., 2003). These methods are very popular in machine learning applications, computer science (data structures), botany (classification), and psychology (decision theory) and are also used as prognostic models in medicine. Nowadays, logistic regression models are used commonly with the purpose of determining risk factors in medical researches and diagnose. In last a few years CTs are attractive because they provide a symbolic representation that lends itself to easy interpretation by humans (Abu-Hanna and de Keizer, 2003, Breiman et al., 1984, Fu, 2004, Kline et al., 2003 and Robnik-Sikonja et al., 2003). The aim of this study is to examine logistic regression and CT methods comparatively in term of results obtained. In direction of this purpose, summarized theoretical explanations belong to both two methods were made and results obtained by examining effects of some social-demographic features on postpartum depression with these methods were compared controversially.
نتیجه گیری انگلیسی
In this study, social-demographic risk factors of postpartum depression occurred in women after delivery by using CT and logistic regression model. As it is determined in other some researchers, it is seen that the postpartum depression risk increases particularly depending on the time after two months period after in this study, too (Buğdaycı et al., 2004 and Heh and Fu, 2003). In addition, it was observed that depression risk is parallel with the increasing of the women’s ages, the increasing in the women’s education levels and the increasing in women’s husbands’ education levels increases depression risk in the women who married early, if the ages at the moment of giving birth are middle age or more. In diagnosing studies, using more than one variable together will increase the diagnose success. In these kinds of researches, CT method gives successful results in term of evaluating these variables together and bringing up relations between variables (Abu-Hanna and de Keizer, 2003, Kline et al., 2003 and Olaru and Wehenkel, 2003).