Article Per Year (5 Year)
p-Index From 2021 - 2026
2.307
P-Index
This Author published in this journals
All Journal JSM (Jurnal SIFO Mikroskil) IJISTECH (International Journal Of Information System & Technology) JSR : Jaringan Sistem Informasi Robotik Jurnal Sistem Komputer & Kecerdasan Buatan Innovation in Research of Informatics (INNOVATICS) IJISTECH JoMMiT : Jurnal Multi Media dan IT DECODE: Jurnal Pendidikan Teknologi Informasi JUSTIN (Jurnal Sistem dan Teknologi Informasi) Jurnal Sains Informatika Terapan (JSIT) Jurnal abimana (Jurnal Pengabdian Kepada Masyarakat Nasional) Jurnal Indonesia Mengabdi
Dani Rofianto
Politeknik Negeri Lampung
Agriculture, Biological Sciences & Forestry Biochemistry, Genetics & Molecular Biology Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Environmental Science Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Materials Science & Nanotechnology Social Sciences Other

Published : 15 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 15 Documents
Search

 1   2 
Performance Evaluation Of SVM With Parameter Optimization On Credit Card Fraud Data Subset Using SMOTE Mahardika, Ahmad Farrel; Fahrezi, Irza Nuzul; Alleredha, Muhammad Hadya; Amaliah, Khusnatul; Rofianto, Dani
IJISTECH (International Journal of Information System and Technology) Vol 9, No 1 (2025): The June Edition
Publisher : Sekolah Tinggi Ilmu Komputer (STIKOM) Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/ijistech.v9i1.398

Abstract

This study evaluates the performance of the Support Vector Machine (SVM) algorithm in detecting credit card fraud by overcoming the class imbalance problem using the Synthetic Minority Oversampling Technique (SMOTE) technique and parameter optimization through Grid Search. The dataset used is sourced from Kaggle, consists of 10,001 transactions, and has been balanced. SMOTE is applied exclusively to the training data to prevent data leakage. The optimization process produces the best parameters at a value of C = 10 and gamma = 0.1. Model evaluation is carried out using recall, precision, F1-score, and AUC-ROC metrics. The results show a significant increase in performance in recognizing fraudulent transactions. The final model recorded a recall of 0.68, precision 0.90, F1-score 0.77, and AUC-ROC 0.98. These findings prove that the combination of SMOTE techniques and parameter optimization can improve the effectiveness of SVM in classifying minority classes more accurately. This approach is considered to have great potential to be applied in automated fraud detection systems in the financial sector.
Prediksi Kekambuhan Kanker Tiroid Menggunakan Algoritma Random Forest Safitri, Egi; Rofianto, Dani; Karnila, Sri; Nurjoko, Nurjoko; Kurniawan, Hendra; Arkhiansyah, Yuni; Rizal, Ruki
Jurnal SISKOM-KB (Sistem Komputer dan Kecerdasan Buatan) Vol. 8 No. 3 (2025): Volume VIII - Nomor 3 - Mei 2025
Publisher : Teknik Informatika, Sistem Informasi dan Teknik Elektro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47970/siskom-kb.v8i3.833

Abstract

Kekambuhan kanker tiroid pasca terapi Radioactive Iodine (RAI) merupakan tantangan penting dalam penatalaksanaan jangka panjang pasien. Penelitian ini bertujuan membangun model prediktif untuk mengidentifikasi potensi kekambuhan dengan memanfaatkan data klinis dan patologis menggunakan algoritma Random Forest. Dataset terdiri atas 383 data pasien dengan 13 atribut, termasuk usia, jenis kelamin, staging kanker, jenis patologi, klasifikasi risiko, dan respons terhadap terapi. Proses pra-pemrosesan meliputi penyandian data kategorik, eksplorasi fitur, dan pembagian data latih dan uji secara stratifikasi. Hasil evaluasi menunjukkan performa tinggi dari model, dengan akurasi 96,5%, presisi 96,7%, recall 90,6%, dan AUC 0,99. Analisis fitur menggunakan SHAP mengungkap bahwa Stage, Response, dan Risk merupakan faktor paling berkontribusi terhadap prediksi kekambuhan. Penelitian ini menunjukkan bahwa model Random Forest tidak hanya efektif dalam klasifikasi biner, tetapi juga dapat diinterpretasikan secara klinis untuk mendukung pengambilan keputusan medis yang lebih personal dan preventif.
Analysis of Climate Change on Agricultural Yields with Principal Component Analysis and Linear Regression Approaches Safitri, Egi; Rofianto, Dani
JUSTIN (Jurnal Sistem dan Teknologi Informasi) Vol 13, No 3 (2025)
Publisher : Jurusan Informatika Universitas Tanjungpura

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26418/justin.v13i3.91434

Abstract

Climate Change has become a global issue that significantly impacts various sectors, including agriculture. This study aims to analyze the influence of climate variables, such as average temperature, rainfall, carbon dioxide (CO2) emissions, and extreme weather events on agricultural yields. The Principal Component Analysis (PCA) method was used to identify the main variability patterns in the data. At the same time, multiple linear regression was applied to determine the relationship between climate variables and crop yields. The analysis showed that temperature and precipitation were the main factors affecting agricultural yields, with increases in temperature being negatively correlated to crop productivity. PCA identified two principal components that explained the variability in the data, while multiple linear regression showed that temperature and extreme weather events significantly affected crop yields. The results underscore the potential of adaptation strategies in the agricultural sector, such as using climate-resilient crop varieties, water resource management efficiency, and agricultural technology innovation to increase resilience to climate change.
Penerapan Algoritma Random Forest dalam Prediksi Emosi Musik Berdasarkan Karakteristik Fitur Audio Spotify Marsya, Nabila Defany; Munadhil, Muhammad Mufrih; Dzakyananta, M Alvin; Amaliah, Khusnatul; Rofianto, Dani
Jurnal Sains Informatika Terapan Vol. 4 No. 2 (2025): Jurnal Sains Informatika Terapan (Juni, 2025)
Publisher : Riset Sinergi Indonesia (RISINDO)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62357/jsit.v4i2.648

Abstract

Emotion classification in music is a crucial aspect in developing context-aware recommendation systems that respond to the listener’s mood. The Random Forest algorithm is used to map song emotions based on Spotify audio features, namely valence, energy, loudness, and danceability, which reflect the psychological and acoustic aspects of music. The dataset was collected through the Spotify Web API and public repositories, consisting of songs released between 2009 and 2019. Data processing involved normalization and labeling emotions into five categories: Angry, Calm, Happy, Neutral, and Sad. The model was trained using 70% of the data and tested with the remaining 30%. Evaluation results showed an accuracy of 98.75%, with perfect F1-scores for the Happy and Sad categories. Valence and energy were found to be the most influential features, while Calm was often confused with Neutral due to similar acoustic patterns. These findings demonstrate that the Random Forest approach is effective in accurately and consistently classifying music emotions based on audio features.
Model Ensembel untuk Deteksi Depresi di Twitter Berbahasa Indonesia Fitri, Melisa; Nurkhotimah, Jihan Susan; Ihsan, Faiz Nurfaadhil; Amaliah, Khusnatul; Rofianto, Dani
Jurnal Sifo Mikroskil Vol. 26 No. 2 (2025): JSM VOLUME 26 NOMOR 2 TAHUN 2025
Publisher : Fakultas Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55601/jsm.v26i2.1812

Abstract

Pentingnya Deteksi dini gangguan kesehatan mental khususnya depresi di era digital saat ini di mana individu lebih cenderung mengekspresikan kondisi emosionalnya melalui media sosial. Penelitian ini bertujuan untuk mengembangkan model ensembel Machine Learning dalam mendeteksi gejala depresi pada postingan media sosial berbahasa Indonesia, khususnya dari platform Twitter. Dataset yang digunakan adalah Depression and Anxiety in Twitter (ID) yang terdiri dari 6.980 teks berlabel. Proses preprocessing mencakup pembersihan data, vektorisasi dengan TF-IDF, dan pemisahan data menggunakan metode overfitting. Empat algoritma yaitu Support Vector Machine, Naïve Bayes, Random Forest, dan AdaBoost dikombinasikan menggunakan Voting Classifier dengan pendekatan soft voting. Evaluasi model dilakukan menggunakan metrik akurasi, precision, recall, dan F1-score, serta visualisasi heatmap korelasi dan learning curve. Hasil menunjukkan bahwa model Voting Classifier menghasilkan kinerja terbaik dengan F1-score makro sebesar 0,996, menunjukkan bahwa pendekatan ensembel efektif dalam meningkatkan akurasi dan stabilitas klasifikasi. Penelitian ini berkontribusi dalam pengembangan sistem deteksi dini gangguan mental berbasis teks bahasa Indonesia yang dapat digunakan oleh lembaga kesehatan, institusi pendidikan, dan organisasi sosial.
Co-Authors - Nurjoko Alleredha, Muhammad Hadya Amaliah, Khusnatul Astria Hijriani, Astria Dewi Kania Widyawati Dzakyananta, M Alvin Eko Win Kenali Fahrezi, Irza Nuzul Fathurrahman Kurniawan Ikhsan Fitra, Jaka Fitri, Melisa Halim Fathoni Halim Fathoni Hendra Kurniawan Ihsan, Faiz Nurfaadhil K Ikshan, Fathurrahman Mahardika, Ahmad Farrel Marsya, Nabila Defany Munadhil, Muhammad Mufrih Neni Purwati Nurkhotimah, Jihan Susan Oki Arifin Rizal, Ruki Safitri, Egi Sri Karnila Tiara Kurnia Khoerunnisa Yuni Arkhiansyah