MODEL PENGAMBILAN KEPUTUSAN DALAM PREDIKSI KASUS TUBERKULOSIS MENGGUNAKAN REGRESI LOGISTIK BERBASIS BACKWARD ELIMINATION

Ratih Sari Wardani(1*), Purwanto -(2)


(1) 
(2) 
(*) Corresponding Author

Abstract


The purpose of this study is to obtain a predictive model in cases of Tuberculosis (TB) using alogistic regression based on backward elimination. The study is conducted through literaturereview, collecting secondary data of 200 patients of Pulmonary Tuberculosis aged more than 15years.  Data is collected from BKPM Semarang, Central Java. The data consist ofcharacteristics, anamnesis, physical examination, laboratory test results data and radiologicalexamination. Logistic regression based  on backward elimination model has better accuracythan logistic regression without attribute selection.
Kata kunci : Tuberkulosis, Regresi Logistik, Backward Elimination, Prediksi

Full Text:

PDF

Article Metrics

Abstract view : 336 times
PDF - 122 times

Refbacks

  • There are currently no refbacks.


UNIMUS | Universitas Muhammadiyah Semarang
Jl. Kedungmundu Raya No. 18 Semarang

email:Info@unimus.ac.id  http://unimus.ac.id