dc.contributor.advisor | Suharjo, Budi | |
dc.contributor.advisor | Sumarno, Hadi | |
dc.contributor.author | Daud, Ibnu | |
dc.date.accessioned | 2023-08-09T02:18:50Z | |
dc.date.available | 2023-08-09T02:18:50Z | |
dc.date.issued | 2023 | |
dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/123301 | |
dc.description.abstract | Penelitian menggunakan data demography dari perusahaan kreditur online. Masalahnya adalah menemukan kriteria market yang tepat, sehingga diperlukan analisis untuk memprediksi variabel yang memengaruhi status calon kreditur menerima pinjaman. Metode yang digunakan adalah supervised learning: random forest classifier, k-NN models, dan logistic linear regression. Dari exploratory data didapatkan bahwa 50,1% adalah kreditur yang berhasil mendapatakn pinjaman dan sisanya adalah Non-ACC Credit sebesar 49,9%. 59,9% calon kreditur berlatar pendidikan sekolah menengah, 20,1% lulusan S1, 14% mereka yang menempuh jenjang diploma, 3,8% adalah pasca-sarjana, dan sisanya 2,1% lulusan sekolah dasar. Status pernikahan didapat 49,1% adalah calon kreditur belum menikah, 48,8% sudah menikah, 1,2% sudah/sedang bercerai, dan sisanya 0,8% adalah janda/duda karena kematian. Untuk membandingkan model menggunakan Confusion Matrix dan didapat bahwa random forest classifier mempunyai nilai persentase terbaik dari semua parameter accuracy, sensitivity, precision, dan F1 score masing-masing adalah 51%. Sedangkan k-NN dan logistic linear regression masing-masing seebesar 50% untuk setiap parameter tersebut. | id |
dc.description.abstract | The research uses demographic data from online creditor companies. The problem is finding market criteria, so analysis is needed to predict the variables that affect the status of prospective creditors receiving loans. The methods are Random Forest Classifier, k-NN Models, and Logistic Linear Regression. Exploratory data shows that 50,1% are creditors who have successfully obtained loans and 49,9% are Non- ACC Credit. 59,9% of prospective creditors have a high school education background, 20.1% have a bachelor, 14% have a diploma, 3,8% postgraduate, and 2,1% have elementary school. Marital status obtained 49,1% single, 48,8% married, 1,2% divorced/divorced, and the remaining 0,8% widow/widower due to death. To compare the models using the Confusion Matrix and it was found that the Random Forest Classifier has the best percentage value of all parameters Accuracy, Sensitivity, Precision, and F1 Score each is 51%. Meanwhile, k-NN and Logistic Linear Regression are 50% each for each of these parameters. | id |
dc.language.iso | id | id |
dc.publisher | IPB University | id |
dc.title | Pemodelan Permintaan Market Kredit Online | id |
dc.title.alternative | Modeling of Demand Market Credit Online | id |
dc.type | Undergraduate Thesis | id |
dc.subject.keyword | accuracy | id |
dc.subject.keyword | data handling | id |
dc.subject.keyword | F1 score | id |
dc.subject.keyword | k-NN models | id |
dc.subject.keyword | logistic linear regression | id |
dc.subject.keyword | precision | id |
dc.subject.keyword | random forest classifier | id |
dc.subject.keyword | sensitivity | id |