Text
IMPLEMENTASI BOOTSTRAP AGGREGATING PADA METODE DECISION TREE DAN REGRESI LOGISTIK UNTUK KLASIFIKASI KANKER SERVIKS
Cervical cancer is one of the most common diseases suffered by women in the world. Global Burden Cancer recorded the incidence of cervical cancer worldwide in 2020 as many as 604,127 cases while the death rate from cervical cancer was recorded at 341,831 cases. The high mortality rate from cervical cancer is also related to the delay in diagnosis of the disease. Therefore, it is necessary to have a study that discusses the classification process to diagnose cervical cancer patients with high accuracy results. The purpose of this study is to classify cervical cancer based on pap smear cell image extraction using the decision tree method and logistic regression with the implementation of bootstrap aggregating (bagging) and without the implementation of bagging. The data used in this study is a dataset of Pap smear cell image extraction of cervical cancer 7 classes originating from Herlev University Hospital. The results of this study indicate that the implementation of bagging can improve the performance of a single method for cervical cancer classification. Classification using the decision tree method resulted in accuracy, precision, recall, and specificity of 85.87%, 56.09%, 54.88%, and 91.48%, respectively. While the classification using the decision tree method with bagging resulted in accuracy, precision, recall, and specificity of 87.89%, 61.85%, 61.59%, and 92.61%, respectively. Classification using logistic regression method resulted in accuracy, precision, recall, and specificity of 89.75%, 68.61%, 68.45%, and 93.75%, respectively. While the classification using the logistic regression method with bagging produces the best accuracy, precision, recall, and specificity, which are 90.53%, 70.13%, 70.70%, and 94.25%, respectively.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2207000932 | T68420 | T684202022 | Central Library (Referens) | Available but not for loan - Not for Loan |
No other version available