TY  - JOUR
TI  - Comparison of Performances of Associative ClassificationMethods for Cervical Cancer Prediction: Observational Study
AB  - Objective: Associative classification is a method thatgenerates a rule-based classifier in a categorical data set. The mainpurpose of the associative classification is to create classificationmodels with high performance and, in addition, to improve inter pretability thanks to the rules it creates. In this study, it is aimed toclassify, predict cervical cancer with the methods of relational clas sification and to determine the most important parameters and rela tional rules associated with the disease. Material and Methods: Inthe study, regular class association rules (RCAR) and classificationbased on associations (CBA) methods were applied to the openaccess data set named “Cervical Cancer Behavioral Risk Data Set”and the results were compared. In order to separate the numericalvariables in the data set, Boruta feature selection method was ap plied to determine the most important features about Ameva andcervical cancer. The performances of the created relational classifi cation models were evaluated with accuracy, balanced accuracy,sensitivity, specificity, Matthews correlation coefficient (MCC), G mean, diagnostic accuracy, Youden’s index, positive predictivevalue, negative predictive value and F1-score criteria. Results: Ac cording to CBA model results, sensitivity is 100%, specificity 98%,accuracy 98.6%, balanced accuracy 99%, Youden's index 98%,MCC 96.7%, diagnostic accuracy 98.6%, G-mean 97.7%, negativepredictive value 1%, positive predictive value 95.5%, and F1 score97.7%. According to RCAR model results, sensitivity is 90.5%,specificity 98%, accuracy 95.8%, balanced accuracy 94.3%,Youden's index 88.5%, MCC 89.8%, diagnostic accuracy 95.8%,G-mean 95.6%, negative predictive value 96.2%, positive predic tive value 95%, and F1 score 92.7%. Conclusion: When the resultsare examined, it can be said that the CBA model is more successfulin classifying cervical cancer compared to the RCAR model. Inaddition, the relational classification models created in this studyand the rules obtained regarding the disease are promising in termsof their use in early diagnosis and preventive medicine practices forcervical cancer.
AU  - YAĞIN, Fatma Hilal
AU  - ARSLAN, Ahmet Kadir
AU  - ÇOLAK, Cemil
AU  - Yağın, Burak
DO  - 10.5336/biostatic.2021-84349
PY  - 2021
JO  - Türkiye Klinikleri Biyoistatistik Dergisi
VL  - 13
IS  - 3
SN  - 1308-7894
SP  - 266
EP  - 272
DB  - TRDizin
UR  - http://search/yayin/detay/503924
ER  -