Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Indonesian Journal on Computing (Indo-JC)

Implicit Boundary Integral Method for Homogeneous Hele-Shaw Problem with multi-connected Domain irma palupi
Indonesia Journal on Computing (Indo-JC) Vol. 4 No. 1 (2019): Maret, 2019
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/INDOJC.2019.4.1.279

Abstract

In this work, we implement the implicit boundary integral method for a homogeneous Hele-Shaw problem with a multi-connected domain. This method base on the solution of layer potential integral for the Laplace equation. The numerical technique is easy to implement, base on the idea of averaging the parameterization near the boundary and applying the Coarea formula. This technique changes the boundary integral into the Riemann integral that numerically easy to compute. The difficulty in the computation of hypersingular integral occurs to compute the normal velocity of free boundary. We use a collocation technique to eliminate the hypersingular part in the integral equation. Also, we show the numerical results and its computation performance due to the appearance of a non-invertible matrix.
Penerapan Analisis Klaster untuk Seleksi Aset dalam Optimasi Portofolio Investasi Saham varid vaya yusuf; irma palupi; indwiarti indwiarti
Indonesia Journal on Computing (Indo-JC) Vol. 5 No. 2 (2020): September, 2020
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34818/INDOJC.2020.5.2.438

Abstract

Manfaat diversifikasi dapat dioptimalkan dengan mengategorikan aset ke dalam kelas-kelas tertentu. Di dalam pasar keuangan, terdapat struktur hirarki antar saham dan dapat dianalisis dengan mengobservasi serangkaian harga saham yang saling berkorelasi. Penelitian terdahulu banyak berfokus pada dampak analisis klaster terhadap performa portofolio, namun sedikit yang meninjau sisi seleksi aset dalam benchmark-nya. Penelitian ini mengajukan tiga skenario alternatif seleksi aset untuk proses konstruksi portofolio berbasis klaster sebagai sudut pandang baru dalam penyusunan benchmark konstruksi portofolio. Dalam pelaksanaannya, digunakan ward’s method untuk melakukan klasterisasi terhadap saham berdasarkan data in-sample dari 606 perusahaan tercatat di BEI. Dilanjutkan dengan konstruksi portofolio dengan tangency portfolio sebagai preferensi portofolio optimal dan seleksi aset dengan tiga skenario alternatif. Performa portofolio diukur menggunakan rasio Sharpe dan rasio terhadap data out-sample. Analisis klaster yang dilakukan menunjukkan kualitas yang luar biasa dalam kelompokkelompok saham yang terbentuk. Portofolio dengan analisis klaster memberikan performa yang sangat baik, melebihi portofolio tanpa analisis klaster.
Lung Cancer Prediction Model using Logistic Linear Regression with Imbalanced Dataset Priscilia Lovita Paelongan; Irma Palupi
Indonesia Journal on Computing (Indo-JC) Vol. 7 No. 2 (2022): August, 2022
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34818/INDOJC.2022.7.2.616

Abstract

Cancer is one of the leading causes of death worldwide. Cancer cases in Indonesia have now reached 4.8 million in 2018. Most cases are breast, cervix, and lung. Furthermore, we need to note that 43 percent of these cancer cases are preventable. This study uses a linear logistics regression model. Linear logistic regression models can be used for categoric datasets. The appropriate model is obtained after parameter assessment, test the significance of each affecting attribute, and test the suitability of the model. This is done to obtain prediction models and risk factors at the level of correlation of disease size. This method is relatively easy and conceptually practical, so it is possible to apply it to diagnose early symptoms of lung cancer. The results include a linear logistics regression model for early prediction of lung cancer patients based on symptoms, habits, and history of health diseases to see the likelihood that someone with a certain level of risk could have lung cancer. The factors that affect a person with lung cancer are difficulty swallowing, coughing, chronic diseases, fatigue, and age.