cover
Contact Name
Hendra Kurniawan
Contact Email
hendra.kurniawan@darmajaya.ac.id
Phone
-
Journal Mail Official
jodmapps@darmajaya.ac.id
Editorial Address
Jl. Z.A. Pagar Alam No. 93 Gedong Meneng, Bandar Lampung Lampung
Location
Kota bandar lampung,
Lampung
INDONESIA
Journal of Data Science Methods and Applications
ISSN : -     EISSN : 30905605     DOI : https://doi.org/10.30873/jodmapps
Theoretical Foundations: Architecture, Management and Process for Data Science Artificial Intelligence Classification and Clustering Data Pre-Processing, Sampling and Reduction Deep Learning Educational Data Mining Forecasting High Performance Computing for Data Analytics Learning Classifiers Learning Theory Optimization Methods Probabilistic and Statistical Models and Theories Scientific Data and Big Data Analytics Statistical Learning Machine Learning and Knowledge Discovery: Big Data Visualization, Modeling and Analytics Data and Knowledge Visualization Database Technology Knowledge Based Neural Networks Knowledge Discovery (Heterogeneous, Unstructured and Multimedia Data) Knowledge Discovery in Network and Link Data Knowledge Discovery in Social Networks Learning for Streaming Data Machine Learning for High-Performance Computing Multimedia/Stream/Text/Visual Analytics Spatial/Temporal Data Computational Data Science: Big Data Computational for Big Data Analysis Computational Intelligence for Pattern Recognition and Medical Imaging Computer Application for Data Analytics Computer Architecture for Data Analytics Computer Graphics for Data Analytics Data Acquisition, Integration, Cleaning Data Visualizations Data Wrangling Databases Decision Making from İnsights, Hidden Patterns Intelligent Information Retrieval Optimization for Data Analytics Probabilistic And İnformation-Theoretic Methods Search and Mining Support Vector Machines Time Series Analysis Applications: Bioinformatics Applications Biomedical Informatics Applications Biometrics Applications Collaborative Filtering Applications Data and Information Semantics Applications Data Mining Algorithms Applications Data Mining Systems Applications Data Streams Mining Applications Database and Information System Performance Applications Database Systems & Applications Electronic Commerce and Web Technologies Applications Electronic Government & E-participation Applications Graph Mining Applications Healthcare Applications Image Analysis Applications Information Retrieval Applications Multimedia Data Mining Applications Natural Language Processing Applications Pre-Processing Techniques Applications Spatial Data Mining Applications Statistical and Scientific Databases Applications Web Search Applications
Articles 7 Documents
Search results for , issue "Vol. 1 No. 2 (2025)" : 7 Documents clear
Perbandingan Performa Model Naïve Bayes dan Regresi Logistik dalam Klasifikasi Kecanduan Media Sosial pada Siswa Nurjoko; Agung, Agung Dwi Praditya; Triyari, Novita; Rafly, M. Rafly Octa; Agus Rahardi
Journal of Data Science Methods and Applications Vol. 1 No. 2 (2025)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

Kecanduan media sosial di kalangan pelajar menjadi isu yang semakin relevan seiring meningkatnya penggunaanplatform digital dalam kehidupan sehari-hari. Penelitian ini bertujuan untuk membangun model klasifikasi gunamemprediksi tingkat kecanduan media sosial berdasarkan data survei siswa. Dataset yang digunakan terdiri dari 705responden dengan 13 atribut yang mencakup aspek demografis, akademik, kebiasaan digital, dan kondisi psikososial.Dua model machine learning yang digunakan dalam penelitian ini adalah Gaussian Naïve Bayes dan Regresi Logistik.Setelah melalui proses data preparation, analisis dan evaluasi model menggunakan data uji sebesar 20%, diperolehhasil bahwa Regresi Logistik memiliki performa yang lebih unggul dengan akurasi mencapai 98%, jauh di atas NaïveBayes yang hanya mencapai 69%. Regresi Logistik juga menunjukkan keseimbangan metrik yang baik, termasukprecision, recall, dan nilai ROC AUC sebesar 0,98. Temuan ini mengindikasikan bahwa Regresi Logistik lebih efektifdan sesuai untuk mengidentifikasi siswa yang berisiko mengalami kecanduan media sosial.
Prediksi Kanker Payudara di Indonesia menggunakan Algoritma Support Vector Machine dan Regresi Logistik Zoe, Zeo Enitisya; ray, raynaldo syah pratama; Prita, Prita Yana Faliha; Keyla, Keyla Artaliani; Della, Rahma Della Mustika Ayu
Journal of Data Science Methods and Applications Vol. 1 No. 2 (2025)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

Breast cancer is one of the leading causes of death among women worldwide, including in Indonesia. Early detection and risk factor analysis are crucial in the efforts toward more effective prevention and treatment. This study aims to identify risk patterns of breast cancer through the analysis of clinical patient data. The dataset includes several clinical variables such as age, menopausal status, tumor size, cancer grade, the number of affected lymph nodes, and hormone status (progesterone and estrogen). The analytical method used is descriptive and exploratory to uncover common patterns within the data. The results indicate that age, menopausal status, and tumor size show a significant correlation with the level of breast cancer risk. These findings are expected to contribute to the development of decision support systems in breast cancer diagnosis and promote greater public awareness of the importance of early detection and regular breast health monitoring.
Pemanfaatan Machine Learning untuk Prediksi Kepuasan Pelanggan pada UMKM Digital Agus Rahardi; Nursalim, Nursalim; Andini, Rekha Aprilia; Tri, Anugrah Tri Agil S; Gilang, Gilang Ramadhan; Dwi, Dwi Salim; Nurjoko
Journal of Data Science Methods and Applications Vol. 1 No. 2 (2025)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

UMKM digital memainkan peran penting dalam perekonomian Indonesia, namun mempertahankan kepuasan pelanggan tetap menjadi tantangan utama. Penelitian ini bertujuan untuk membangun model prediksi kepuasan pelanggan menggunakan algoritma Machine Learning seperti Logistic Regression, Decision Tree, dan Random Forest. Dataset simulasi sebanyak 10.000 entri pelanggan digunakan, mencakup fitur-fitur seperti frekuensi pembelian, nilai transaksi rata-rata, rating layanan, dan metode pembayaran. Model dievaluasi berdasarkan metrik B. Hasil penelitian menunjukkan bahwa algoritma Random Forest memberikan akurasi dan kinerja terbaik dalam mengklasifikasikan kepuasan pelanggan. Temuan ini menunjukkan potensi besar penggunaan Machine Learning dalam membantu UMKM digital meningkatkan kualitas layanan dan loyalitas pelanggan.
Segmentasi Pelanggan Berdasarkan Kebutuhan Primer Skunder dan Tersier Menggunakan K-Means Clustering Indah, Caesaliana Indah Mu’assyaroh; Arkan, M. Rizieq Sultan; Galuh, Galuh Sitoresmi; Sabrina, Sabrina Rizkiya; Zida, Zida Nadhifah Aulia Kencana
Journal of Data Science Methods and Applications Vol. 1 No. 2 (2025)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

In the digital marketing era, companies are required to deeply understand customer behavior in order to develop targeted strategies. Customer segmentation is a common technique used to group customers based on similarities in their characteristics and consumption behaviors. This study aims to identify customer segments using unsupervised learning techniques with the K-Means clustering algorithm. The dataset, obtained from Kaggle, contains 2,240 customer records with demographic and purchase behavior attributes. The six primary features analyzed include Income, Age, TotalChildren, MntMeatProducts, NumCatalogPurchases, and Recency. The clustering results reveal distinct customer groups with different characteristics and purchasing tendencies, which can be used to develop more personalized and efficient marketing strategies.
Prediksi Tingkat Stres Mahasiswa Selama Pembelajaran Daring Menggunakan Algoritma Machine Learning Rocky, Rocky Khalifah Akbar; dede; Aldo, Aldo Septian Raharjo; Celvin, Celvin Immanuel Suhendar
Journal of Data Science Methods and Applications Vol. 1 No. 2 (2025)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

The sudden transition to online learning during the COVID-19 pandemic has had a significant psychological impacton students, particularly in the form of increased stress levels. This study aims toidentify and analyze the factors that influence student stress during online learningusing a quantitative approach and predictive modeling. Data were obtained from 100 students aged 18–25 years, covering variables such as screen time, sleep duration, physical activity, pre-exam anxiety, and changes inacademic performance. Statistical analysis showed that high screen time, less than 6 hours of sleep, andacademic anxiety were significantly associated with increased stress levels (p < 0.01). The Random Forest modelsuccessfully predicted stress categories with 82% accuracy and identified sleep duration as the mostdominant factor. These findings indicate the need for more adaptive academic policy reforms regarding mental health,including digital load management, healthy sleep education, and the integration of psychological support. This studyprovides an empirical basis for educational institutions to design data-driven preventive interventions toreduce the prevalence of stress among students.
Prediksi Jenis Ancaman Siber Global Menggunakan Algoritma Random Forest Nanda P, Yudista; Ali, Rionaldi; Aris, Noki; Saputra, Agus; Saputra, M . David
Journal of Data Science Methods and Applications Vol. 1 No. 2 (2025)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

Global cyber threats continue to increase along with the widespread digital transformation across various sectors. This study aims to predict the types of cyberattacks based on global historical data from 2015 to 2024. Data was obtained from the Global Cybersecurity Threats dataset, which includes information on countries, affected sectors, and types of attacks. The method used was supervised learning with the Random Forest algorithm, which is known to be effective for classifying and analyzing complex variables. The results show that this algorithm is capable of identifying attack patterns with high accuracy and assisting in early threat detection. This research is expected to contribute to the development of data-driven cybersecurity systems and predictive modeling. Global cyberthreats continue to grow in complexity, along with society's increasing reliance on digital systems. This study aims to analyze trends and predict the types of cyberthreats based on historical data from 2015 to 2024, obtained from the Global Cybersecurity Threats dataset. The method used was supervised learning with the Random Forest algorithm to classify attack types and predict potential financial losses. The analysis results show that the model can identify important patterns that can assist organizations in mitigating cyber risks. This research contributes to the development of data-driven cyber threat intelligence systems
Prediksi Pengunduran Diri Karyawan Menggunakan Metode Algoritma Random Forest Prasetyo, Bima Restu; Apiliani, Lusy Pebi; Intan, Citra Nur; Jonathan, Kenny
Journal of Data Science Methods and Applications Vol. 1 No. 2 (2025)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

Employee attrition is a critical issue in human resource management as it directly affects a company’s productivity and operational efficiency. Therefore, a data-driven prediction system is needed to identify potential employee resignation risks at an early stage. This study aims to build an employee attrition classification model using the Random Forest algorithm, implemented in the RapidMiner software. The dataset used in this study is derived from the IBM HR Analytics Employee Attrition Dataset. The research process includes data cleaning, attribute transformation, model building, and performance evaluation using a confusion matrix and metrics such as accuracy, precision, and recall. The results show that the Random Forest model achieved an accuracy of 91.04%, a precision of 100% for the “Yes” class, and a recall of 44.37%. Furthermore, it was found that the variables JobLevel and TotalWorkingYears significantly influence attrition status. Therefore, this model can serve as a decision support tool in identifying employee attrition risks and designing more effective, data-driven retention strategies

Page 1 of 1 | Total Record : 7