Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : jurnal teknik informatika dan sistem informasi

Pengembangan Aplikasi Maranatha E-signature Bernard Renaldy Suteja; Mewati Ayub; Kafka Febianto Agiharta
Jurnal Teknik Informatika dan Sistem Informasi Vol 9 No 1 (2023): JuTISI (in progress)
Publisher : Maranatha University Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28932/jutisi.v9i1.5657

Abstract

Using digital signature in a document, especially in work from home situation, becomes crucial to misuse. The digital signature is a schema to identify somebody uniquely and to prove the originality of a document. The development of information technology allows digital signatures as a tool to ensure the originality of information. The aim of the study is to develop an application that uses Quick Response (QR) Code as a digital signature. So QR Code can work as a tool to authenticate the leader’s signature or document verification. After the application was built and applied, users’ evaluation was performed through questionnaires. The result shows that 44% of users state that the application has been well used and beneficial. There are some suggestions for application revision from 34% of users, whereas 22% have no comment.  
Perbandingan Akurasi Model Pembelajaran Mesin untuk Prediksi Seleksi Masuk Perguruan Tinggi Negeri Oktavianus Yopi Wardana; Mewati Ayub; Andreas Widjaja
Jurnal Teknik Informatika dan Sistem Informasi Vol 9 No 1 (2023): JuTISI (in progress)
Publisher : Maranatha University Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28932/jutisi.v9i1.6126

Abstract

Seleksi Nasional Masuk Perguruan Tinggi Negeri (SNMPTN) is still one of the favorite admission routes for high school students to continue their education at Perguruan Tinggi Negeri (PTN). SNMPTN uses semester 1 to 5 report card scores for 6 subjects that are inputted in Pangkalan Data Sekolah dan Siswa (PDSS). Prediction of SNMPTN can be done using machine learning models with various methods. This study aims to create a predictive model using the Decision Tree CART, Gaussian Naïve Bayes and Logistic Regression methods, make predictions and compare the level of accuracy of the models made. The methodology used in this research is Knowledge Discovery in Database (KDD). This is to get useful knowledge from data. The dataset used is data on the scores of 6 subjects for 5 semesters from class 2015 to 2022. Model evaluation uses the Split Percentage Method and K-Fold Cross Validation. The results show that the accuracy scores for the 3 models are different. Logistic Regression has a score of 0.82, followed by Decision Tree CART with a score of 0.75 and finally Gaussian Naïve Bayes with a score of 0.70. The hypothesis put forward by the researcher is in accordance with the results obtained, that the Logistic Regression model has a higher accuracy score. Mathematically, Logistic Regression is not too complicated when compared to other models. To get a model that fits with needs must involve iterating through the machine learning process and trying various variations.