| dc.contributor.advisor | Ardana, Ngakan Komang Kutha | |
| dc.contributor.advisor | Sumarno, Hadi | |
| dc.contributor.author | Al Ubaidah, Hafidz | |
| dc.date.accessioned | 2024-08-24T05:28:15Z | |
| dc.date.available | 2024-08-24T05:28:15Z | |
| dc.date.issued | 2024 | |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/158459 | |
| dc.description.abstract | Alat kardiotokografi mengklasifikasikan kesehatan janin ibu hamil dengan
21 faktor ke dalam 3 klasifikasi kesehatan yaitu sehat, terjangkit, dan patologi.
Analisis komponen utama digunakan untuk mereduksi dimensi data kesehatan janin
ibu hamil tanpa kehilangan informasi penting di dalamnya. Keberadaan komponen
utama membuat klasifikasi kurang efektif sehinga diperlukan seleksi variabel.
Salah satu pendekatan seleksi variabel adalah pemilihan variabel berdasarkan nilai
loading absolut. Sementara Regresi Logistik Multinomial digunakan untuk
memodelkan klasifikasi kesehatan janin. Hasilnya diperoleh 6 komponen utama
dan 6 variabel yang signifikan terhadap kesehatan janin yaitu gerakan janin,
deselerasi parah, variabilitas jangka pendek abnormal, minimum histogram FHR,
maksimum histogram FHR, dan kecenderungan histogram dengan hasil evaluasi
model F1 score terboboti mencapai 0.78 dan AUC terboboti sebesar 0.87. | |
| dc.description.abstract | Cardiotocography measurements classify the fetal health of pregnant
women with 21 factors into 3 health classifications healthy, affected, and pathology.
Principal Component Analysis is used to reduce the dimensionality of pregnant fetal
health data without losing important information. The existence of principal
components make the classification less effective so that variable selection is
needed. One of the variable selection approaches is the selection of variables based
on the absolute loading value on. Multinomial Logistic Regression was used to
model the classification of fetal health. The results obtained 6 principal components
and 6 variables that are significant to fetal health, namely fetal movement, severe
deceleration, abnormal short term variability, minimum FHR histogram, maximum
FHR histogram, and histogram tendency with the results of model evaluation F1
score weighted to 0.78 and AUC weighted to 0.87. | |
| dc.description.sponsorship | | |
| dc.language.iso | id | |
| dc.publisher | IPB University | id |
| dc.title | Klasifikasi Kesehatan Janin dengan Regresi Logistik Multinomial menggunakan Seleksi Variabel pada Analisis Komponen Utama | id |
| dc.title.alternative | | |
| dc.type | Skripsi | |
| dc.subject.keyword | Principal Component Analysis (PCA) | id |
| dc.subject.keyword | classification | id |
| dc.subject.keyword | Multinomial Logistics Regression | id |
| dc.subject.keyword | fetal health | id |
| dc.subject.keyword | variable selection | id |