Claim Missing Document
Check
Articles

Found 2 Documents
Search

PROTOTIPE SISTEM PENDETEKSI HUJAN BERBASIS ARDUINO UNO MENGGUNAKAN SENSOR RAIN YL-83 Akhlis Munazilin; Islamiyatul Addewiyah; Fitra Lia Shofiah
Jurnal Riset Teknik Komputer Vol. 3 No. 2 (2026): Juni : Jurnal Riset Teknik Komputer (JURTIKOM)
Publisher : CV. Denasya Smart Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69714/54rfkd66

Abstract

Rain detection systems are essential in tropical countries such as Indonesia, where sudden rainfall can occur and cause significant losses including wet laundry, localized flooding, and disruptions to outdoor activities. This research aims to design and implement an Arduino UNO-based rain detection prototype utilizing the YL-83 Rain Sensor module. The system employs an ATmega328P microcontroller to process analog signals from the rain sensor through its 10-bit Analog-to-Digital Converter (ADC). When the sensor's ADC reading is less than or equal to 400, indicating the presence of rainwater on the sensor surface, the system activates a buzzer alarm and illuminates a red LED indicator. Conversely, when the ADC value exceeds 400, indicating dry conditions, the buzzer is deactivated and a green LED is illuminated. The circuit was assembled on a breadboard following a Fritzing schematic, with the sensor's AO pin connected to Arduino's A0 analog input, buzzer to digital pin 7, red LED to pin 9, and green LED to pin 8, each LED using a 220-ohm current-limiting resistor. Testing across 15 data points demonstrated 100% detection accuracy in distinguishing rain from non-rain conditions, with ADC values ranging from 230–400 for rain conditions and 501–1023 for dry conditions. The system demonstrates reliable, low-cost performance as a rain detection alert mechanism and offers a strong foundation for future IoT-enabled enhancements such as wireless notification via ESP8266/ESP32 integration..
Klasifikasi Data Penyakit Demam Berdarah Dengue (DBD) Melalui Algoritma Decision Tree Dengan RapidMiner Fitra Lia Shofiah; Zaehol Fatah
Jurnal Mahasiswa Teknik Informatika Vol. 5 No. 1 (2026): Volume 5 Nomor 1 April 2026
Publisher : Universitas Ngudi Waluyo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35473/jamastika.v5i1.4571

Abstract

Demam Berdarah Dengue (DBD) merupakan permasalahan kesehatan masyarakat yang hingga kini terus mendapat perhatian serius di Indonesia. Proses diagnosis untuk DBD biasanya dilakukan melalui pemeriksaan laboratorium dan evaluasi klinis, Akan tetapi, pendekatan konvensional ini cenderung membutuhkan waktu yang relatif panjang serta sangat bergantung pada analisis profesional kesehatan. Oleh karena itu, diperlukan suatu pendekatan berbasis teknologi yang mampu mempercepat sekaligus mempermudah proses klasifikasi secara lebih akurat. Penelitian ini bertujuan untuk mengidentifikasi tingkat keparahan penyakit Demam Berdarah Dengue (DBD) dengan menerapkan algoritma Decision Tree yang dijalankan melalui aplikasi RapidMiner. Data yang digunakan merupakan algoritma sekunder yang di ambil dari internet,berisi 200 catatan pasien dengan delapan artibut medis, yaitu suhu tubuh, jumlah trombosit, jumlah leukosit, tekanan darah, ruam kulit, sakit kepala, nyeri otot, dan muntah. Langkah-langkah dalam penelitian mencakup pengumpulan data, preprocessing, penerapan algoritma Decision Tree, serta evaluasi model dengan memanfaatkan confusion matrix dengan metode cross-validation. Temuan dari studi ini memperlihatkan bahwa model Decision Tree mencapai tingkat ketepatan akurasi mencapai 85%, precision 84%, recall 83%, dan F1-score 83%, dengan trombosit dan leukosit sebagai variabel yang paling dominan dalam proses klasiifikasi. Dengan merujuk pada hasil yang diperoleh, studi ini menyimpulkan bahwasanya algoritma Decision Tree cukup efektif dalam mengklasifikasikan tingkat keparahan penyakit DBD dan dapat di gunakan sebagai dasar untuk mengembangkan system pendukung keputusan di bidang kesehatan. Kata Kunci: Demam Berdarah Dengue, Data Mining, Decision Tree,RapidMIner, Klasifikasi   Dengue Hemorrhagic Fever (DHF) is a public health problem that continues to receive serious attention in Indonesia. The diagnosis process for DHF is usually carried out through laboratory examinations and clinical evaluations. However, this conventional approach tends to take a relatively long time and relies heavily on the analysis of health professionals. Therefore, a technology-based approach is needed that can accelerate and simplify the classification process more accurately. This study aims to identify the severity of Dengue Hemorrhagic Fever (DHF) by applying the Decision Tree algorithm run through the RapidMiner application. The data used is a secondary algorithm taken from the internet, containing 200 patient records with eight medical attributes, namely body temperature, platelet count, leukocyte count, blood pressure, skin rash, headache, muscle pain, and vomiting. The steps in the study include data collection, preprocessing, application of the Decision Tree algorithm, and model evaluation using the confusion matrix with the cross-validation method. The findings of this study show that the Decision Tree model achieved an accuracy rate of 85%, precision of 84%, recall of 83%, and an F1-score of 83%, with platelets and leukocytes as the most dominant variables in the classification process. Based on these results, this study concludes that the Decision Tree algorithm is quite effective in classifying the severity of dengue fever and can be used as a basis for developing decision support systems in the healthcare sector. Keywords: Dengue Hemorrhagic Fever, Data Mining, Decision Tree, RapidMiner, Classification