JSI (Jurnal Sistem Informasi) Universitas Suryadarma

UI/UX Design of a Mobile-Based Application for Private Tutoring Services (Online and Offline) in Samarinda City Using the User Centered Design Method

2026-02-19T15:21:55+08:00

Samarinda City has numerous tutoring services and private tutors spread across various areas. However, the community still faces difficulties in finding information regarding costs, programs, and tutor options that suit their needs. Therefore, it is necessary to design a mobile application capable of displaying details of learning programs, costs, tutor selection, and scheduling for both online and offline learning to facilitate the community. This application is named SamaLearn (Samarinda Learn), which aims to help students solve problems outside of school, understand subjects considered difficult, improve social skills, and encourage academic achievement. Its design follows the User Centered Design method through the stages of Understand Context of Use, Specify User Requirements, Design Solutions, and Evaluation Against Requirements. The test results using usability testing and the System Usability Scale yielded a score of 87.5, categorized as acceptable, with a grade scale B and an adjective rating of excellent. These findings prove that the UI/UX design of the SamaLearn application has met user needs.

Application of the Double Diamond Method to Produce an Engaging Mobile Application for Faznet Samarinda

2026-02-19T15:30:10+08:00

Faznet Samarinda is a local internet service provider offering internet access to its customers. To enhance the user experience, a more attractive, intuitive, and functional mobile application design is required. This study aims to redesign the UI/UX of the Faznet application according to user needs using the Double Diamond method, which consists of the discover, define, develop, and deliver stages. Data were collected through literature studies, in-depth interviews, and observation of user comments on the Google Play Store. The outcome of this research is a new UI/UX design evaluated using usability testing with Maze and the System Usability Scale (SUS) questionnaire. Testing involving 20 respondents yielded a Maze score of 95 and an SUS score of 93, falling into the "Excellent" category with a "Grade A". These results indicate that the redesign successfully meets user needs and resolves previous issues.

Quality of Service Analysis of WLAN Networks using the PPDIOO Method at the Department of Dental Health, Poltekkes Tasikmalaya

2026-05-05T10:18:13+08:00

This study aims to evaluate the quality of WLAN network services at the Department of Dental Health, Poltekkes Tasikmalaya, which has a bandwidth of 150 Mbps but frequently experiences connectivity issues. The method used is PPDIOO (Prepare, Plan, Design, Implement, Operate, Optimize) with Quality of Service (QoS) parameters based on the TIPHON standard. Measurements were conducted at 15 locations using the nPerf and Wireshark applications. Observation results indicate several critical points, such as the Auditorium and the D4 Faculty Room, fall into the "Poor" category with packet loss reaching 89%. As a solution, a proposed topology was designed through Cisco Packet Tracer simulation, which includes the addition of access points and account-based bandwidth management. The simulation results show an improvement in service quality, reaching the "Good" to "Excellent" categories.

Analisis Sentimen Komentar Tiktok Terhadap Kenaikan Harga Plastik Menggunakan Algoritma Naive Bayes

2026-05-12T12:56:48+08:00

Kebijakan pengurangan limbah plastik melalui peningkatan harga di Indonesia menimbulkan beragam persepsi masyarakat, terutama di platform TikTok.Studi ini difokuskan pada pembedahan opini pengguna TikTok terkait kebijakan penyesuaian harga plastik dengan mengimplementasikan metode Naive Bayes, sekaligus mengukur seberapa presisi model tersebut dalam memetakan pandangan publik. Data yang digunakan terdiri dari 398 komentar yang dikumpulkan melalui proses crawling dan pembersihan data, kemudian dianalisis melalui ekstraksi kata kunci dan visualisasi word cloud. Hasil evaluasi menunjukkan bahwa model mencapai akurasi sebesar 88,2% dengan AUC 0,885 dan MCC 0,812, menunjukkan performa yang sangat baik dalam memprediksi sentimen. Hasil analisis menunjukkan mayoritas komentar bersifat netral, mencerminkan bahwa masyarakat masih dalam tahap observasi terhadap kebijakan tersebut, namun harga plastik sering dikaitkan dengan kenaikan biaya kebutuhan pokok lain seperti BBM dan minyak goreng. Temuan ini penting untuk pengambil kebijakan agar dapat memahami persepsi masyarakat dan mengarahkan strategi komunikasi serta kebijakan yang lebih efektif.

Perancangan Sistem Aplikasi Penyimpanan Digital Laporan On The Job Training Siswa Pemandu Lalu Lintas Udara Di Indonesia Aviation School

2026-05-20T10:46:37+08:00

On The Job Training merupakan kegiatan pembelajaran yang dirancang untuk dapat menurunkan siswa secara langsung untuk bekerja agar mencapai tingkat keahlian tertentu dan menyinkronkan antara program pendidikan yang ada di sekolah dengan program keahlian yang diperoleh melaui kegiatan On The Job Training. Pengalaman On The Job Training yang telah dimilki siswa tidak akan berarti apa-apa tanpa adanya dorongan dari siswa untuk berprestasi. penyimpanan buku laporan On The Job Training yang belum terkomputerisasi sehingga memakan banyak tempat dan juga tidak tersimpan dengan rapi, selain itu dengan laporan On The Job Training yang berupa buku (fisik) dapat beresiko hilang dan rusak serta sulitnya untuk mencari suatu laporan On The Job Training yang telah disimpan karena harus mencari satu per satu. Untuk itu , penulis berinisiatif membuat sistem penyimpanan arsip digital laporan On The Job Training yang terintegrasi dengan database yang berbasis web dengan menggunakan bahasa pemograman PHP. Metode yang digunakan dalam penelitian ini adalah studi pustaka , studi lapangan, dan wawancara. Sistem penyimpanan buku laporan On The Job Training dengan berbasis web ini dapat mengatasi masalah yang ada sehingga menjadi lebih efektif dan tersimpan aman dalam database

Classification of Telecommunication Customer Churn Using Logistic Regression and Support Vector Machine

2026-06-02T11:49:28+08:00

Customer churn is a critical issue in the telecommunication industry as it directly affects a company's revenue. This study aims to develop and compare Logistic Regression and Support Vector Machine (SVM) models for predicting customer churn using the Telco Customer Churn dataset from IBM Watson Analytics, which consists of 7,043 customer records. The research process includes data exploration, data preprocessing, model training, and evaluation using Stratified K-Fold Cross-Validation (k = 5). The experimental results show that Logistic Regression achieved an accuracy of 80.70% with an average cross-validation score of 0.8043, while SVM achieved an accuracy of 79.28% with an average cross-validation score of 0.7954. Feature analysis indicates that tenure, MonthlyCharges, contract type, and internet service type are the most influential factors affecting customer churn. Based on these results, Logistic Regression demonstrates superior and more stable performance in predicting telecommunication customer churn.

Analisis Komparatif Determinan Prestasi Matematika dan Bahasa Indonesia Menggunakan Random Forest dan SHAP

2026-06-20T13:06:11+08:00

This study aims to compare the determinants of academic performance between Mathematics and Indonesian Language subjects using an Educational Data Mining approach. Utilizing the UCI Student Performance dataset, comprising 395 Mathematics and 649 Indonesian Language students, we identified 382 students who were enrolled in both subjects through identity-based merging. Random Forest models were optimized via Grid Search and evaluated using RMSE, MAE, and R², while SHAP (SHapley Additive exPlanations) was employed for model interpretability. The results indicate that the Random Forest model achieved higher predictive accuracy for Indonesian Language (R² = 0.85) compared to Mathematics (R² = 0.78). Feature importance and SHAP analyses revealed significant differential impacts: Mathematics performance was predominantly influenced by absences (r = -0.28), study time (r = 0.31), and weekend alcohol consumption (r = -0.22), whereas Indonesian Language performance was more strongly determined by maternal (r = 0.32) and paternal (r = 0.28) education levels. We conclude that educational intervention strategies should be tailored specifically to each subject rather than applying a generalized approach.

Predicting Inflation Rates in Indonesia Using Linear Regression Algorithms

2026-06-23T09:25:24+08:00

Inflasi merupakan indikator ekonomi makro yang berpengaruh signifikan terhadap stabilitas perekonomian nasional. Penelitian ini bertujuan untuk memprediksi tingkat inflasi bulanan di Indonesia menggunakan algoritma regresi linier berbasis platform Orange Data Mining. Metode yang digunakan adalah pendekatan kuantitatif dengan eksperimen komputasional, memanfaatkan dataset sebanyak 72 observasi bulanan yang memuat dua variabel prediktor, yaitu Bulan_Ke dan Siklus_Bulan, serta satu variabel target berupa Inflasi. Sebanyak 64 observasi digunakan sebagai data pelatihan dan delapan observasi sisanya sebagai data pengujian. Model dikonfigurasi tanpa regularisasi dengan intercept aktif. Hasil prediksi menunjukkan nilai inflasi berkisar antara -0,1078% hingga 0,4982%, dengan nilai tertinggi pada siklus bulan ke-12 dan terendah pada siklus bulan ke-8. Simpulan penelitian ini menunjukkan bahwa algoritma regresi linier mampu menghasilkan prediksi yang interpretatif dengan mempertimbangkan pola musiman melalui variabel Siklus_Bulan dan tren jangka panjang melalui variabel Bulan_Ke.

Klasifikasi Sentimen Komentar TikTok Terkait Kenaikan Harga BBM 2026 di Sumatera Selatan Menggunakan Algoritma Naive Bayes pada Orange Data Mining

2026-06-23T09:33:08+08:00

Penyesuaian subsidi harga Bahan Bakar Minyak (BBM) di Indonesia memicu respons publik yang luas, termasuk di wilayah Sumatera Selatan. Penelitian ini menganalisis sentimen masyarakat Sumatera Selatan terhadap kenaikan harga BBM menggunakan data komentar TikTok. Sebanyak 476 komentar yang telah melalui tahap prapemrosesan diklasifikasikan ke dalam tiga kategori: negatif, netral, dan positif. Distribusi data terdiri dari 127 instans negatif, 243 instans netral, dan 106 instans positif. Proses klasifikasi dilakukan dengan algoritma Naive Bayes Classifier melalui framework Orange Data Mining. Evaluasi model menggunakan k-fold cross-validation menunjukkan nilai Area Under Curve (AUC), Accuracy, F1-score, Precision, dan Recall sebesar 1.000. Analisis teks menggunakan word cloud dan pembobotan kata menunjukkan bahwa istilah “bbm”, “harga”, “pertalite”, “pertamax”, dan “subsidi” merupakan kata kunci dominan dalam diskusi publik. Hasil penelitian ini mengindikasikan bahwa pendekatan pemrosesan bahasa alami berbasis Naive Bayes dapat digunakan untuk memantau persepsi publik terhadap kebijakan energi di tingkat regional.

Mapping Educational Achievement Inequality Between Provinces in Indonesia Using K-Means Clustering with Davies-Bouldin Validation and Silhouette Score

2026-06-24T17:26:07+08:00

The disparity in education quality between regions remains a real challenge in Indonesia, necessitating accurate regional mapping as a basis for targeted policymaking. This study aims to cluster 38 provinces in Indonesia based on educational achievement indicators to identify regions with extreme disparities. The method used is the K-Means Clustering algorithm with attribute selection and Z-Score-based data normalization stages executed using RapidMiner software. Validation of the optimal number of clusters was carried out using the Elbow Method approach and dual evaluation metrics, namely the Davies-Bouldin Index (DBI) and Silhouette Score. The results showed that the 3-cluster configuration was the most optimal, with the lowest DBI value (0.58) and the highest Silhouette Score (0.78). This clustering successfully identified two New Autonomous Regions (DOB) in Papua, namely the Papua Mountains Province and the Papua Central Province, which are included in the low (critical) education quality cluster with a very extreme gap in educational ability and illiteracy rates compared to other regions. In conclusion, the application of the K-Means algorithm based on comprehensive validation has proven effective in highlighting regional priorities, so that it can become an empirical basis for the government in its efforts to equalize educational facilities specifically in Indonesia.

Analysis of Factors Influencing the Choice of College Majors for Vocational High School Students Using a Linear Regression Approach

2026-06-23T09:28:08+08:00

Pemilihan jurusan di perguruan tinggi merupakan keputusan krusial bagi lulusan SMK yang menentukan arah karier dan kehidupan profesional mereka. Penelitian ini bertujuan mengidentifikasi faktor-faktor dominan yang memengaruhi keputusan siswa kelas XII SMK Pembangunan Jaya Yakapi Pasar Minggu dalam memilih jurusan perguruan tinggi. Metode yang digunakan adalah survei deskriptif kuantitatif dengan instrumen kuesioner 30 butir yang disebarkan kepada 150 responden. Analisis data menggunakan statistik deskriptif, uji validitas, reliabilitas, korelasi Spearman, dan regresi linear berganda. Hasil penelitian menunjukkan bahwa minat pribadi (mean=4.32) dan prospek kerja (mean=4.18) merupakan faktor paling dominan. Keenam variabel secara bersama-sama menjelaskan 71.4% variasi keputusan pemilihan jurusan (R²=0.714, F=61.83, p<0.001). Temuan ini diharapkan menjadi acuan bagi sekolah dan perguruan tinggi dalam merancang layanan bimbingan karier yang lebih efektif bagi lulusan SMK.