Claim Missing Document
Check
Articles

Found 3 Documents
Search

Digital Transformation Impact: Pengaruh Digitalisasi terhadap Kinerja UMKM Putri Intan Prastiwi; Arif Setiawan; Zaenal Mustofa; Nova Mei Indriyani; Nadia Putri
JURNAL ILMIAH EDUNOMIKA Vol. 10 No. 2 (2026): EDUNOMIKA
Publisher : ITB AAS Indonesia Surakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29040/jie.v10i2.19931

Abstract

This study aims to analyze the effect of digital transformation on the performance of Micro, Small, and Medium Enterprises (MSMEs) in Surakarta, particularly in the fashion, culinary, and creative economy sectors. The study examines four dimensions of digital transformation, namely business process transformation, business model transformation, domain transformation, and organizational/cultural transformation, and their influence on MSME performance. This research employed a quantitative approach using a survey method. Data were collected through questionnaires distributed to 50 MSME owners selected using a non-probability sampling technique with purposive sampling method. Data analysis was conducted using Structural Equation Modeling–Partial Least Square (SEM-PLS). The results indicate that business process transformation, business model transformation, domain transformation, and organizational/cultural transformation have a positive and significant effect on MSME performance. Business process transformation was identified as the most dominant variable influencing MSME performance through operational digitalization, administrative automation, and technological integration. The findings demonstrate that digital transformation plays a strategic role in improving sales growth, productivity, profitability, and business competitiveness of MSMEs. The novelty of this study lies in the integration of four dimensions of digital transformation in analyzing MSME performance within the context of fashion, culinary, and creative economy sectors in Surakarta.
Implementation of the Threshold Method in an IoT-Based Monitoring System for Temperature, Humidity and Feed in Broiler Chicken Houses Febryan Djastin Arya Raharja; Joni Maulindar; Arif Setiawan
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 3 (2026): Research Paper July 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i3.8808

Abstract

Broiler chickens require optimal house management to maintain productivity, particularly in terms of temperature, humidity, and feed availability. Monitoring that is still done manually can potentially cause delays in detecting changes in environmental conditions and feed stock, which can hinder the chickens’ growth. This study aims to implement a threshold method in an Internet of Things (IoT)-based monitoring system for temperature, humidity, and feed levels in broiler chicken houses. The study employed the ADDIE (Analysis, Design, Development, Implementation, Evaluation) development model, utilizing an ESP32 microcontroller, a DHT22 sensor to measure temperature and humidity, an HC-SR04 sensor to detect feed levels, a 4-channel relay module to control fans and incandescent lights, and a web dashboard connected to a Supabase database for real-time monitoring. The system applies threshold values adjusted to the broiler chickens’ growth phases so that the actuators can operate automatically to maintain optimal coop conditions. Test results show that the system is capable of monitoring temperature, humidity, and feed levels in real time, transmitting data to the web dashboard, and controlling lights and fans according to threshold values specifically, lights turn on when the temperature is below 31°C and fans turn on when the temperature is above 31°C. The implementation of this system improves the effectiveness of monitoring and controlling coop conditions and has the potential to support poultry farm productivity. The system applies threshold values adjusted to the growth phase of the broiler chickens so that the actuators can operate automatically to maintain optimal conditions in the coop. Test results show that the system is capable of monitoring temperature, humidity, and feed levels in real time, transmitting data to a web dashboard, and controlling lights and fans according to threshold values specifically, lights turn on when the temperature is below 31°C and fans turn on when the temperature is above 31°C. The implementation of this system improves the effectiveness of monitoring and controlling coop conditions and has the potential to support broiler farm productivity. Further development can be pursued by adding automatic notification features, improving sensor accuracy, and integrating an automatic feeding system.
Analisis Perbandingan Algoritma Oversampling: Smote Dan Adasyn Pada Random Forest Untuk Prediksi Stroke Musa Muhammad Isa Al’Azzam; Afu Ichsan Pradana; Arif Setiawan
Prosiding Seminar Nasional Teknologi Informasi dan Bisnis Prosiding Seminar Nasional Teknologi Informasi dan Bisnis (SENATIB) 2026
Publisher : Fakultas Ilmu Komputer Universitas Duta Bangsa Surakarta

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

Abstract

Stroke merupakan penyebab utama kematian dan kecacatan secara global, menjadikan deteksi dini berbasis rekam medis sebagai tindakan preventif yang krusial. Dalam memodelkan data klinis, algoritma Machine Learning seringkali menghadapi masalah ketimpangan distribusi kelas (class imbalance) yang parah. Pada dataset stroke publik, ketidakseimbangan ini menyebabkan model klasifikasi terjebak pada accuracy paradox, di mana model memperoleh akurasi tinggi dengan cara memprediksi seluruh data sebagai kelas mayoritas (sehat), namun sama sekali gagal mendeteksi kelas minoritas (stroke). Penelitian ini secara komprehensif membandingkan efektivitas dua algoritma penyeimbangan data level sintetis, yaitu Synthetic Minority Over-sampling Technique (SMOTE) dan Adaptive Synthetic (ADASYN), yang diintegrasikan pada arsitektur Random Forest. Dataset yang digunakan adalah Healthcare Stroke Data (Kaggle) dengan komposisi 95,13% pasien sehat dan 4,87% pasien stroke (5.110 observasi). Metodologi penelitian mencakup pra-pemrosesan (imputasi median dan label encoding), penerapan oversampling eksklusif pada data latih, dan pengujian menggunakan data tak terlihat (unseen data). Hasil eksperimen menunjukkan bahwa model Baseline (tanpa oversampling) mencatat akurasi semu sebesar 95,11% namun memiliki metrik Recall yang sangat kritis (2,00%). Sebaliknya, integrasi algoritma SMOTE terbukti paling efektif memecahkan paradoks tersebut dengan mendongkrak Recall hingga 11 kali lipat menjadi 22,00% serta mencapai F1-Score 0,1864, meskipun menurunkan akurasi keseluruhan menjadi 90,61%. ADASYN menghasilkan kinerja yang sedikit lebih rendah dengan Recall 18,00% akibat sintesis data agresif pada wilayah perbatasan (decision boundary) yang menciptakan noise. Kesimpulan dari penelitian ini menegaskan bahwa penyeimbangan data secara merata melalui SMOTE sangat vital dalam sistem diagnostik medis yang memprioritaskan sensitivitas penemuan penyakit (Recall).