Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : INSTALL: Information System and Technology Journal

Exploratory Data Analysis (EDA) of Marriage Patterns in Kabupaten Banjar Using Machine Learning Approaches Husna Karima; Mambang; Subhan Panji Cipta; Muhammad Zulfadhilah
INSTALL: Information System and Technology Journal Vol 1 No 2 (2024): INSTALL : Information System and Technology Journal
Publisher : LPPM Universitas Sari Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33859/install.v1i2.629

Abstract

Marriage is a sacred moment that has a significant impact on the social, economic and demographic structure of a region. This research aims to implement a marriage dataset in Banjar Regency and find a correlation between the number of marriages, education level and age of the bride and groom using Exploratory Data Analysis (EDA) techniques and machine learning approaches. The method used is a quantitative method with observation and analysis using EDA and machine learning. The research results show that there is a strong correlation between the number of marriages and the age of the bride and groom (r = 0.99) and between the number of marriages and the education level of the bride and groom (r = 0.99). In addition, a perfect correlation was found between the ages of the groom and the bride (r= 0.99) as well as between the educational levels of the groom and the bride (r = 1). This analysis provides a better understanding of marriage patterns in Banjar Regency and shows that couples aged 21-30 years have a high positive correlation with the number of marriages. It is hoped that these results can become the basis for social policies and educational programs related to marriage.
Analysis of the Utilization of TikTok Content as a Coping Strategy to Reduce Stress Among Final-Year Students Using a Classification Method Husna Karima; Zulfadhilah, Muhammad; Prastya, Septyan Eka; Pratiwi, Evi Lestari
INSTALL: Information System and Technology Journal Vol 2 No 3 (2025): INSTALL : Information System and Technology Journal
Publisher : LPPM Universitas Sari Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33859/install.v2i3.991

Abstract

Stress represents a prevalent psychological challenge among final- year university students, particularly during thesis completion. Academic pressure, social demands, and future uncertainty trigger stress that negatively impacts mental health. Social media, especially TikTok, is increasingly utilized as a coping mechanism to reduce stress through entertainment, educational, and motivational content. This study aims to analyze TikTok content utilization as a coping strategy for stress reduction among final-year students using a classification method. This quantitative research employed a survey approach with a population of 342 active TikTok users among final- year students at Sari Mulia University. Data were collected through an online questionnaire covering variables including content type, duration, features used, and psychological indicators such as anxiety, emotions, escapism, and coping effectiveness. Data preprocessing included one-hot encoding, SMOTE, and normalization, followed by classification using Support Vector Machine with RBF kernel optimized through GridSearchCV. Results revealed very high correlations among psychological variables (r ≈ 0.93–1.00), while correlations between content type and stress reduction were relatively low (0.00–0.15). Some pure entertainment content showed negative correlations with psychological improvement. The SVM model achieved high classification accuracy of approximately 94%. This study demonstrates that TikTok can serve as a short-term stress coping tool for final-year students, though its effectiveness depends heavily on the type of content consumed. Educational and motivational content shows greater potential for stress reduction compared to pure entertainment content. This research contributes to understanding digital mental health support mechanisms and provides insights for developing healthier media consumption strategies among university students.
The Effectiveness of early stopping on the efficiency of training CNN models for phishing URL identification Rifani, Muhammad Rifani; Prastya, Septyan Eka; Zulfadhilah, Muhammad; Munsyi
INSTALL: Information System and Technology Journal Vol 3 No 1 (2026): INSTALL : Information System and Technology Journal
Publisher : LPPM Universitas Sari Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33859/install.v3i1.1024

Abstract

Phishing is a significant cybersecurity threat in which malicious URLs deceive users to steal sensitive data. Traditional detection methods, such as blacklists, often fail to keep pace with evolving phishing techniques. Deep learning, particularly Convolutional Neural Networks (CNNs), offers strong potential in phishing URL classification by capturing structural and semantic character-level patterns. However, CNN training demands high computational resources and risks overfitting. This study investigates the effectiveness of early stopping as a regularization technique to improve efficiency and generalization in character-based CNN models. Using a large-scale dataset of 130.080 URLs across four classes (benign, phishing, malware, defacement), the model employed character tokenization, embedding, convolution-pooling layers, and softmax classification. Early stopping monitored validation loss with patience values of 3, 5, and 10 epochs. Results show a 51% training time reduction and accuracy improvement from 96% to 97%, confirming early stopping as an efficient and robust detection approach.
Co-Authors ., Mambang Abdul Kadir Adryan Ramadhan Ahmad Busairi Ahmad Faisal Ahmad Ghazali Madhony Ahmad Riki Renaldi Ahmad Riki Renaldy Angga Irawan Anggraini Susfarwanti Annisa Annisa Anshori Prasetya, Muhammad Riko Antonia Yenitia Asyiah Asyiah Aulia Rahma Aulia, Hudatul AULIA, RIZKA Bayu Nugraha Bayu Nugraha Bima Wicaksono Cipta, Subhan Panji Darini Kurniawati Dewi Pusparani Sinambela, Dewi Pusparani Dwi Salmarini, Desilestia Eka Prastya, Septyan Ermadiningtyas, Retno Evi Lestari Pratiwi - Politeknik Hasnur Kalimantan Selatan, Evi Lestari Pratiwi Finki Dona Marleny Fitriani Fitriani Gusti Zahratunnisa Hadi, Nofie Haldi Budiman Haniffah Sri Rinjani Heni Pujiastuti Hikmah, Rahmadaniati Hudatul Aulia Husna Karima Husna Karima Imam Riadi Indah Wulandari Irawan, Angga Iwan Yuwindry Jaya Hari Santoso Junius Akbar Karlina Karlina Kartika Kartika Kartika Kartika Kelana, Enisda Libra Lisda Handayani, Lisda Lisyanti, Fatthiya Lufila Fila M Samsul Hasbi M Samsul Hasmi Mambang Maria Ulfah Maulana, Maghfur Maulana, Rahmat Melda Melda Miranda Miranda Misnawati Muhammad Khairul Akbar Muhammad Riduan Syafi’i Muhammad Satrio Ayuba Muhammad Zaini Bakri Muhammad Ziki Elfirman Munsyi Muthia Elma Mutmainah Mutmainah Naparin, Husni Nastiti, Kunti Nita Hestiyana, Nita Noor Pratama, Ramadhani Nopie Hadi Nor Azizah Novalia Widiya Ningrum Novita Dewi Iswandari Nur Hidayah Nur Lathifah Nur Meilianti Maulida Nur Syifa Nurhaeni Nurhaeni Nurhaeni Nurhaeni NURUL HIDAYAH Pebriadi, Muhammad Syahid Prastya, Septyan Eka Putri Putri Putri Yuliantie Rahmini Rahmini Rhafiq Abdul Ghani Riduansyah, Muhammad Rifani, Muhammad Rifani Risma Maulida Risma Risma Rismawati Rismawati Rizkian Muhammad Fikri Ropikah Ropikah Rudy Anshari Samita, Mambang Sandro Nesta Pembriano Sari, Rahmadah Septian Eka Prastya Septyan Eka Prasetya Septyan Eka Prastya Septyan Eka Prastya Setia Budi Shopa Handayani Siti Gadis Hardianti Subhan Panji Cipta Subhan Panji Cipta Sultan Arrasyid Susanti Suhartati, Susanti Syapotro, Usman Tasya Salsabila Umi hanik Fetriyah, Umi hanik Usman Syapotro Viviana Viviana Winda Maolinda, Winda Wulandari Febriani Wusko, Ikna Urwatul Yudi prayudi Yunandar Yunandar Yusri Yusri Zaini Lambri Assyaifi