Claim Missing Document
Check
Articles

Found 20 Documents
Search

Model Machine Learning Untuk Prediksi Risiko Penyakit Liver Dengan Random Forest Teroptimasi Rizky Andrea Arifa; Nana Suarna; Agus Bahtiar; Nining Rahaningsih; Willy Prihartono
Jurnal Sistem Informasi dan Teknologi Vol 6 No 1 (2026): Jurnal Sistem Informasi dan Teknologi (SINTEK)
Publisher : LPPM STMIK KUWERA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56995/sintek.v6i1.204

Abstract

Penyakit liver merupakan salah satu kondisi kronis dengan tingkat mortalitas tinggi, sehingga diperlukan pendekatan prediksi yang akurat untuk mendukung deteksi dini. Penelitian ini bertujuan mengembangkan model machine learning untuk memprediksi risiko penyakit liver menggunakan algoritma Random Forest yang dioptimalkan dengan RandomizedSearchCV. Dataset yang digunakan terdiri dari 1.700 entri yang mencakup variabel klinis dan gaya hidup, termasuk usia, jenis kelamin, BMI, konsumsi alkohol, kebiasaan merokok, riwayat genetik, aktivitas fisik, diabetes, hipertensi, serta hasil Liver Function Test. Proses penelitian meliputi preprocessing, normalisasi skala, pembagian data menggunakan train-test split 80:20, pembangunan model baseline, dan optimasi hiperparameter. Hasil eksperimen menunjukkan bahwa optimasi menghasilkan peningkatan performa model, dengan akurasi 0.91, peningkatan recall sebesar 3.20%, dan AUC-ROC mencapai 0.96. Analisis feature importance menunjukkan bahwa LiverFunctionTest, BMI, dan AlcoholConsumption merupakan fitur paling berpengaruh terhadap prediksi risiko penyakit liver. Dengan demikian, Random Forest teroptimasi terbukti efektif dalam menghasilkan model prediksi yang akurat dan dapat digunakan sebagai alat pendukung keputusan dalam deteksi dini penyakit liver.
ALGORITMA RANDOM FOREST UNTUK PREDIKSI STATUS PINJAMAN BERDASARKAN SKOR KREDIT Hadit Attaufiqqurrohman; Ade Irma Purnamasari; Denni Pratama; Nining Rahaningsih; Willy Prihartono
METHODIKA: Jurnal Teknik Informatika dan Sistem Informasi Vol. 12 No. 1 (2026): Volume 12 Nomor 1 Tahun 2026
Publisher : Universitas Methodist Indonesia

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

Abstract

The rapid development of financial technology has encouraged financial institutions to adopt data-driven credit scoring systems in order to minimize the risk of default. However, many loan eligibility prediction models still face challenges such as data imbalance (class imbalance) and the limited capability of traditional models to capture non-linear relationships among variables. This study aims to develop a loan status prediction model using the Random Forest algorithm combined with the Synthetic Minority Oversampling Technique (SMOTE) and One-Hot Encoding (OHE) to improve model accuracy and generalization capability. The data used in this study are secondary data obtained from the public Kaggle platform, consisting of 45,000 records with 14 demographic and financial attributes. The research method employs a supervised learning approach with several stages, including data acquisition and preprocessing (data cleaning, normalization, encoding, and data balancing), Random Forest model training, and performance evaluation using accuracy, precision, recall, F1-score, and AUC metrics. The results show that the combination of Random Forest, SMOTE, and OHE achieves high predictive performance, with an accuracy of 94.8%, precision of 95.6%, recall of 93.7%, F1-score of 94.6%, and an AUC value of 0.972. The most influential variables in loan status prediction are credit_score, person_income, and loan_amnt. This approach is proven to be effective in addressing data imbalance issues and improving classification accuracy in identifying creditworthy and non-creditworthy borrowers.
Penerapan Algoritma C4.5 untuk Optimalisasi Manajemen Stok Obat di Apotek Nafa Farma Khairunnisa Amarullah; Rini Astuti; Willy Prihartono; Ryan Hamonangan
IJAI (Indonesian Journal of Applied Informatics) Vol 9, No 2 (2025)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20961/ijai.v9i2.96224

Abstract

Abstrak : Manajemen stok obat menjadi tantangan utama di apotek Nafa Farma untuk mencegah kelebihan atau kekurangan stok. Penelitian ini mengaplikasikan algoritma C4.5 sebagai metode klasifikasi untuk mendukung pengelolaan stok yang optimal. Data stok obat dari Januari sampai Desember 2023 dianalsis menggunakan pendekatan Knowledge Discovery in Database (KDD) dengan software RapidMiner. Penelitian ini menunjukkan bahwa algoritma C4.5 dapat meningkatkan efisiensi manajemen stok obat dengan akurasi 80,67% dan F1-score rata-rata 80.52% ini memberikan rekomendasi strategis untuk pengadaan obat. Obat kategori laku direkomendasikan untuk diutamakan dalam pengadaan, sementara obat tidak laku dapat dikurangi pembeliannya untuk menghindari pemborosan. Algoritma C4.5 efektif untuk meningkatkan efisiensi pengelolaan stok obat====================================================Abstract : Drug stock management is a major challenge at Nafa Farma pharmacy to prevent excess or shortage of stock. This research applies the c4.5 algorithm as a classification method to support optimal stock management. Drug stock data from January to December 2023 was analyzed using a Knowledge Discovery in Database (KDD) approach with RapidMiner software. This study shows that the C4.5 algorithm can improve the efficiency of drug stock management with an accuracy of 80.67% and an average F1-score of 80.52%, providing strategic recommendations for drug procurement.  Sellable category drugs are recommended to be prioritized in procurement, while unsellable drugs can be reduced in purchase to avoid waste. The C4.5 algorithm is effective in improving the efficiency of drug stock management.
Comparison of Balancing Strategies for Classifying Guava Fruit Diseases Putri Nabilla; Nana Suarna; Agus Bahtiar; Nining Rahaningsih; Willy Prihartono
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 5 No. 2 (2026): February 2026
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v5i2.1859

Abstract

The problem of class imbalance often poses an obstacle in deep learning-based image classification, especially in the domain of digital agriculture. The imbalance in data distribution makes it easier for models to recognize the majority class, while performance for the minority class declines. This study aims to analyze the effectiveness of three strategies for handling class imbalance: Weighted Loss Function, Oversampling, and a combination of Weighted Loss and Oversampling, in improving the performance of image classification of guava fruit diseases using a transfer learning-based MobileNetV2 architecture. The dataset consists of 3,784 images of three disease classes, namely Anthracnose, Fruit_Fly, and Healthy_guava, which show an imbalanced distribution. The research was conducted through the stages of Exploratory Data Analysis (EDA), pre-processing, augmentation, model training with four scenarios, and evaluation using Accuracy, Precision, Recall, F1-Score, and Macro Average F1-Score. The results showed that the Combination model (Oversampling and Weighted Loss) performed best on the minority class with an F1-score of 0.9630, the highest among all models. The Oversampling strategy produced the highest Macro F1-score of 0.9617, while Weighted Loss provided a significant improvement in classification sensitivity but was still below the combination model. Thus, it can be concluded that the combination strategy is the most effective approach in improving the sensitivity of the model to minority classes, while Oversampling excels in the overall performance stability of the model.
Model Pendampingan Transformasi Digital Terintegrasi Dalam Meningkatkan Kemandirian Dan Daya Saing UMKM Kota Cirebon Bani Nurhakim; Willy Prihartono; Putri Indah Lestari; Revan Faturochman
AMMA : Jurnal Pengabdian Masyarakat Vol. 4 No. 5 : Juni (2025): AMMA : Jurnal Pengabdian Masyarakat
Publisher : CV. Multi Kreasi Media

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

Abstract

Digital transformation has become a strategic necessity for Micro, Small, and Medium Enterprises (MSMEs) as they navigate shifting consumer behaviors, intensifying competition, and the evolution of technology-driven trade ecosystems. However, many MSMEs still face various obstacles, including limited digital literacy, suboptimal use of social media and online marketplaces, weak brand identity, disorganized administration and financial record-keeping, and a lack of integration between marketing, transaction, and customer management processes. This community service initiative aims to implement an integrated digital transformation mentoring model to enhance the self-reliance and competitiveness of MSMEs in Cirebon City. The implementation employs a participatory and needs-based mentoring approach, comprising stages such as needs identification, digital maturity mapping, training, technology implementation, intensive mentoring, monitoring, and evaluation. The mentoring program focuses on five key dimensions: digital branding, digital marketing, digital commerce, digital payments, and digital business management. Results indicate that this phased and integrated mentoring approach successfully improved participants' understanding of digital transformation, their ability to manage business social media, the quality of promotional content, the utilization of digital sales channels and payment methods, and their awareness regarding the importance of business data recording. This model underscores that MSME digital transformation cannot be achieved solely through one-way training; rather, it requires a continuous process involving assessment, hands-on practice, mentoring, evaluation, and follow-up. The program is expected to serve as a replicable and scalable MSME empowerment model, fostered through collaboration among higher education institutions, local government, business communities, and other stakeholders.
Implementasi Algoritma Naïve Bayes untuk Prediksi Penerima Bantuan Sosial di Desa Cigayam Dede Hoeriah; Bani Nurhakim; Sandy Eka Permana; Willy Prihartono; Gifthera Dwilestari
TAMIKA: Jurnal Tugas Akhir Manajemen Informatika & Komputerisasi Akuntansi Vol 4 No 1 (2024): TAMIKA: Jurnal Tugas Akhir Manajemen Informatika & Komputerisasi Akuntansi
Publisher : Universitas Methodist Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46880/tamika.Vol4No1.pp52-58

Abstract

Social assistance is one of the government's programmes aimed at improving the lives of people especially for those who are economically disadvantaged. However, there are several reasons why some people are unable to access social assistance. In the case of this study, the authors used the Naïve Bayes algorithm with the KDD (Knowledge Discovery Database) method to predict the population in obtaining social assistance. The data was taken from the population data of Cigayam Village and the social welfare recipient data in the village ofCigayam with the results showing high accuracy in this study, for the true or false outcome of 1047 data and 53 data with the precision grade of 95.18%, 81.17%, for the real outcome, and 28.38% for the wrong outcome. So with the ROC curve shows the accuracy of the spinning visually, with an AUC of 0.868% for naïve bayes using the ROK curve of 0.90.1.
Analysis of Beverage Sales Data Using the FP-Growth Algorithm at Sini Aja Cafe Widisa Adi Kumara; Rini Astuti; Willy Prihartono; Tati Suprapti
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.772

Abstract

The growth of information technology and data mining techniques has greatly helped analyze consumer purchasing behavior, particularly in marketing and inventory management. This study aims to uncover association patterns between products frequently bought by customers at Sini Aja Cafe and to measure these patterns' support and confidence values. The research uses Knowledge Discovery in Databases (KDD), including stages like data selection, preprocessing, transformation, applying the FP-Growth algorithm, and interpreting results. Data from 1,083 beverage sales transactions at Sini Aja Cafe from August 1 to October 31, 2024. The findings reveal five significant association rules when applying a minimum support of 0.1 (10%) and confidence of 0.3 (30%). Notably, if customers buy Red Velvet Oreo, there is a 56% chance they will also buy Thai Tea. Thai Tea sales dominate with a support value 0.557 (55.7%). The support values of the association rules range from 0.141, categorized as medium, and the confidence values range from 0.235, categorized as low. These findings offer valuable insights for the cafe owner to optimize operations, enhance customer satisfaction, and increase profits.
Website Based Digital Branding Strategy for Increase Sales of Gunung Puntang Coffee In Mekarjaya, Bandung Regency Juliyanti; Rini Astuti; Willy Prihartono
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.802

Abstract

This research aims to develop a branding strategy through optimizing a website-based digital company profile to increase sales of Gunung Puntang coffee. Gunung Puntang coffee is a high-quality local product that requires a digital approach to marketing to reach a broader market and enhance competitiveness. In today's digital era, a website plays a crucial role as a promotional and informational medium, providing customers with easy access to product information and enabling online purchases. The research employs the Prototype method, consisting of problem identification, planning, requirements analysis, system design, implementation, and testing phases. Data collection was conducted through observation, interviews with the coffee business owner, and documentation studies related to business processes and branding strategies. The collected data serves as a basis for designing a website system that optimizes the company's profile and supports coffee sales transactions. The system development includes creating use case diagrams, activity diagrams, and system architecture designs to outline functional and non-functional requirements. The research outcome is a website functioning as a digital information medium for branding Gunung Puntang coffee products and supporting sales transactions. Key features include customer registration, product selection, quantity adjustment, payment methods, order confirmation, and order cancellation. Testing results indicate that the system operates effectively and meets user needs. This website enhances operational efficiency, expands market reach, and improves the shopping experience for customers. It serves as an effective medium for strengthening branding and marketing strategies in the digital era, ensuring the sustainability of local businesses in the global market. Regular evaluations and feature upgrades are recommended to maintain system relevance to customer needs and technological advancements..
Implementation of Naive Bayes in Sentiment Analysis of CapCut App Reviews on the Play Store Oka Alvianto; Willy Prihartono; Fathurrohman
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.805

Abstract

The CapCut video editing application has gained significant popularity among mobile users. This study aims to analyze user sentiment towards CapCut reviews on the Play Store using the Naive Bayes algorithm. User reviews were collected and preprocessed to clean and prepare the text for analysis. The Naive Bayes algorithm was employed to classify the reviews into positive and negative sentiment categories. Findings indicate that the majority of user reviews are positive, highlighting features such as ease of use, attractive visual effects, and the ability to share videos on social media. However, negative reviews were also identified, primarily criticizing issues like bugs, intrusive advertisements, and limitations in specific features. This research provides valuable insights into user sentiment towards CapCut, which can be utilized by developers to enhance application quality and user experience.
Web-Based Chatbot Development and User Satisfaction Analysis Using the Naive Bayes Method Through Online Questionnaires Nurholis; Willy Prihartono; Fathurrohman
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.823

Abstract

This study aims to develop a web-based chatbot using Natural Language Processing (NLP) technology and the Naive Bayes algorithm to enhance digital interaction quality. User satisfaction was evaluated through an online survey involving 202 university students, focusing on ease of use, response speed, and relevance. The research followed the CRISP-DM framework, including data preprocessing (case folding, tokenization, stopword removal, and stemming), text transformation using the TF-IDF method, and implementation of a Naive Bayes classification model. an F1-score of 84%. Sentiment analysis revealed predominantly positive feedback, reflecting user satisfaction with the chatbot’s ease of use and response accuracy. However, some limitations, such as insufficient contextual understanding, were identified. These findings provide valuable insights into NLP-based chatbot development to support effective digital interactions. The proposed chatbot demonstrates potential applications in customer service, education, and e-commerce, with future improvements suggested to enhance contextual comprehension and scalability.