Claim Missing Document
Check
Articles

Found 6 Documents
Search

Penerapan Algoritma Decision Tree dalam Klasifikasi Data Prediksi Kelulusan Mahasiswa M Riski Qisthiano; Putri Armilia Prayesy; Istiana Ruswita
G-Tech: Jurnal Teknologi Terapan Vol 7 No 1 (2023): G-Tech, Vol. 7 No. 1 Januari 2023
Publisher : Universitas Islam Raden Rahmat, Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (398.644 KB) | DOI: 10.33379/gtech.v7i1.1850

Abstract

Dalam melakukan proses klasifikasi terhadap prediksi kelulusan mahasiswa, ada banyak faktor dan kriteria dalam mengukur kelulusan mahasiswa tersebut, serta menentukan mahasiswa tersebut tepat atau tidaknya menyelesaikan studi. Oleh sebab itu, maka dibutuhkan suatu metode klasifikasi untuk melakukan pengukuran terhadap data prediksi kelulusan tepat waktu, penulis menggunakan dataset yang berasal dari beberapa perguruan tinggi tersebar di Kota Palembang. Model yang digunakan ini menggunakan Decision Tree yang berfungsi sebagai salah satu metode untuk melakukan klasifikasi. Dataset yang digunakan adalah data alumni yang sudah dikumpulkan berasal dari perguruan tinggi di Kota Palembang, sedangkan kriteria untuk melakukan poses klasifikasi adalah jurusan, perguruan tinggi setiap mahasiswa, jenis kelas pilihan, dan nilai setiap mahasiswa yang diambil mulai dari semester awal sampai semester ke 4, lalu ada data tahun kelulusan mahasiswa tersebut, beserta data tahun masuk dari mahasiswa tersebut. Setelah peneliti menentukan atribut data yang akan mejadi bagian dari proses klasifikasi, peneliti memilih menggunakan alat bantu Rapidminer dalam mengelola data klasifikasi dengan model decision tree. Proses berikutnya penguji menggunakan 5 kali proses uji K-Fold Validation dengan membagi dataset ke dalam training dan testing. Hasil penelitian ini merupakan akurasi dari hasil klasifikasi terhadap peridiksi yang didapat dari alat bantu Rapidminer dan metode Decision Tree yang memiliki hasil akurasi sebesar 87.93%.
STUDI PERBANDINGAN METODE SUPPORT VECTOR MACHINE, RANDOM FOREST, DAN CONVOLUTIONAL NEURAL NETWORK UNTUK KLASIFIKASI PENYAKIT KULIT Prayesy, Putri Armilia
Jurnal Kecerdasan Buatan dan Teknologi Informasi Vol. 4 No. 1 (2025): January 2025
Publisher : Ninety Media Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69916/jkbti.v4i1.214

Abstract

Penyakit kulit merupakan salah satu masalah kesehatan yang sering terjadi dan membutuhkan diagnosis yang cepat dan akurat untuk menghindari komplikasi atau mempercepat penanganan. Namun, proses diagnosis manual seringkali memakan waktu dan bergantung pada keahlian dokter. Keterlambatan dalam diagnosis dapat menyebabkan perburukan kondisi pasien, memperpanjang waktu pemulihan, dan memperpanjang durasi perawatan atau menyebabkan komplikasi yang lebih serius. Untuk mengatasi permasalahan ini, teknologi pembelajaran mesin dapat dimanfaatkan untuk mengotomatisasi proses klasifikasi penyakit kulit. Penelitian ini membahas perbandingan tiga metode klasifikasi utama Support Vector Machine (SVM), Random Forest, dan Convolutional Neural Network (CNN), untuk menganalisis dataset citra kulit normal dan penyakit kulit. Dataset terdiri dari berbagai jenis kulit yang telah melalui preprocessing data, seperti normalisasi, augmentasi data, dan ekstraksi fitur, guna meningkatkan kualitas data sebelum pelatihan model. Dataset yang digunakan dalam penelitian ini di menggunakan  data science yang bersumber dari kaggle. Hasil penelitian menunjukkan bahwa CNN memberikan performa terbaik dengan akurasi mencapai 92%, berkat kemampuannya menangkap pola non-linear dalam citra. Random Forest menunjukkan performa yang stabil dengan akurasi 85%, terutama pada dataset yang lebih terstruktur. Sementara itu, SVM mencatat akurasi 78%, tetapi memiliki keterbatasan pada data berdimensi tinggi. Kesimpulannya, CNN lebih unggul untuk klasifikasi citra kulit kompleks, sementara Random Forest dan SVM dapat menjadi alternatif untuk dataset sederhana. Penelitian ini memberikan kontribusi dalam pengembangan teknologi AI untuk mendukung diagnosis penyakit kulit yang lebih cepat dan akurat.
Classification of the Fluency Multipurpose of Bank Mandiri Credit Payments Based on Debtor Preferences Using Naive Bayes and Neural Network Method Prayesy, Putri Armilia; Negara, Edi Surya
JOIN (Jurnal Online Informatika) Vol 7 No 1 (2022)
Publisher : Department of Informatics, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

One that has an important role in generating bank profits is providing credit to customers, but credit also carries a very high risk. For this reason, in providing credit to debtors, of course the bank will utilize the personal data of prospective debtors in detail to avoid the risk of problems that will arise in the future. One of the appropriate risks for banks in providing credit is the behavior of customers who do not pay installments at the time which causes bad loans. To overcome and overcome the many bad events, there is an algorithmic calculation method with an intelligent computing system that helps banks in selecting prospective debtors who will be given credit. There are many algorithmic methods that can be used in this kind of research. This study analyzes the classification of staffing credit based on the criteria that become the Bank's standard.The data used by the author in this study uses existing debtor credit data from 2017 to 2020, the modeling process is carried out using split validation with the Naive Bayes algorithm and Neural Network, with this algorithm the 1,314 datasets is divided into 2 parts, namely 80% used as training data and 20% used as testing data. The results showed that the Neural Network algorithm has better results with a correct value of 84.13%, while the Naive Bayes algorithm only produces a value of 72.62%
Web-Based Educational Platform for Diseases and Drugs using a Large Language Model (LLM) Qisthiano, M Riski; Ruswita, Istiana; Prayesy, Putri Armilia
International Journal Scientific and Professional Vol. 5 No. 1 (2026): December 2025 - February 2026
Publisher : Yayasan Rumah Ilmu Professor

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56988/chiprof.v5i1.150

Abstract

This study presents the development of a web-based educational platform leveraging a large language model (LLM) to provide general information on diseases and medications. The platform integrates a curated database of diseases and drugs with the LLM via retrieval-augmented generation. When users enter a disease or drug name, the system retrieves relevant data and uses it as context for the LLM to produce concise responses in a friendly, accessible style. The application is built using native PHP with a MySQL backend and Bootstrap for responsive design. Safety features such as a mandatory disclaimer and filters for emergency conditions ensure that the chatbot does not offer diagnostic or prescriptive advice. Expert reviews indicated that the model-generated content aligned well with the database, and user testing showed high satisfaction with clarity and usability. These results demonstrate that combining structured medical data with a modern LLM can improve public access to reliable health education while maintaining ethical boundaries.
Comparison of Random Forest and SVM Algorithms in Credit Risk Evaluation Based on Debtor Occupation Prayesy, Putri Armilia; Pujakesuma, Angga
Intelligent System and Computation Vol 7 No 2 (2025): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v7i2.431

Abstract

Credit is one of the main sources of income for banking institutions and plays a crucial role in supporting long-term profit growth. However, credit distribution is inherently associated with risks, especially the risk of default when borrowers fail to meet their repayment obligations as agreed. One effective strategy to minimize such risks is to conduct a comprehensive and accurate creditworthiness assessment of prospective borrowers before loan approval is granted. This study aims to evaluate the performance of three classification algorithms—Random Forest, Support Vector Machine (SVM), and Artificial Neural Network (ANN)—in predicting credit risk based on the borrower’s occupation. The dataset used consists of 1,314 loan records with an imbalanced distribution between performing and non-performing loans. The experimental results show that the Random Forest algorithm achieved the highest accuracy at 97%, followed by Support Vector Machine at 73% and Artificial Neural Networks at 64%. While ANN is capable of capturing complex patterns through multilayered learning, Random Forest proved to be the most effective and robust in handling the given dataset. These findings clearly indicate that Random Forest can serve as a reliable method for financial institutions to enhance credit risk evaluation and minimize potential losses arising from loan defaults.
Implementasi User Acceptance Testing (UAT) Pada Pengujian Sistem Informasi Akademik dan Keuangan Santri Putri, Devinka; Afriansyah, Riki; Prayesy, Putri Armilia
TeIKa Vol 15 No 2 (2025): Jurnal
Publisher : Fakultas Teknologi Informasi - Universitas Advent Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36342/ztavmp10

Abstract

The digitization of administration in Islamic boarding schools is an important requirement for improving the effectiveness of academic and financial data management for students. The manual system that was previously used sometimes caused delays in reporting, risks of recording errors, and limited access to information for parents and administrators. Based on these needs, the researcher developed and evaluated a website-based Academic and Financial Information System for students at an Islamic boarding school to support transparency in reporting and ease of access. The system was developed using a prototyping approach utilizing the Laravel framework. The system was tested using the User Acceptance Testing (UAT) method, which involved 31 respondents as direct users. The UAT instrument consisted of 22 questions covering user-friendliness, user satisfaction, system functionality, and system performance. The test results showed that all aspects received a “very good” rating with the following average percentages: ease of use 88.5%, satisfaction 91.05%, functionality 90.07%, and performance 89.6%. These findings confirm that the system has worked according to the operational needs of the Islamic boarding school, is capable of managing academic, financial, and student savings data in an integrated manner, and supports the involvement of guardians through real-time access to information. Then, the system is considered feasible to implement and has the potential to improve the quality of administrative services in Islamic boarding schools.