Claim Missing Document
Check
Articles

Found 2 Documents
Search

Implementasi Sistem Irigasi Berbasis Internet of Things (IoT) Untuk Optimasi Penggunaan Air Pada Pertanian Haj, Shafly Ulya; Adek, Rizal Tjut; Suwanda, Rizki
METIK JURNAL (AKREDITASI SINTA 3) Vol. 9 No. 2 (2025): METIK Jurnal
Publisher : LPPM Universitas Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47002/sqe99d72

Abstract

Traditional irrigation systems in Indonesia's agricultural sector still face obstacles in water use efficiency. This research aims to develop an Internet of things (IoT)-based automatic irrigation system capable of monitoring and controlling soil moisture and water level in real-time. The system is designed using an ESP32 microcontroller, two soil moisture sensors, and one ultrasonic sensor integrated with a servo motor actuator. Rule-based control logic is applied, where the sluice gates open when the moisture is < 50% and the water level is > 25 cm, and close again when the opposite condition occurs. The prototype was tested on two simulated paddy fields with glass aquarium media for three days and five recording cycles. Data was sent to the server and displayed via a web interface for remote monitoring and control. The test results show that the system is able to work responsively, with an average response time of 1.8 seconds, and can increase water use efficiency by 30%. The graph visualisation and manual-automatic control features provide ease of system management. The application of this system shows great potential in supporting the implementation of efficient, scalable, and sustainable precision agriculture.
Location Entity Recognition in Instagram Captions Using Support Vector Machine Algorithm Arifa, Cut Hilma; Adek, Rizal Tjut; Afrillia, Yesy
VOCATECH: Vocational Education and Technology Journal Vol 7, No 1 (2025): August
Publisher : Akademi Komunitas Negeri Aceh Barat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38038/vocatech.v7i1.238

Abstract

AbstractThe rapid advancement of digital technology has significantly influenced productivity and facilitated access to information in daily life, particularly through the widespread use of social media. Instagram is one of the most popular platforms, where text in captions often contains location-related information that can be utilized for spatial analysis. This study aims to identify and classify location entities in Instagram captions using Support Vector Machine algorithm combine with rule-based Named Entity Recognition approach. The method involved linguistic feature extraction based on explicit spatial context, data labeling, model training, and performance evaluation using standard classification metrics: accuracy, precision, recall, and f1-score. Dataset consists of 400 captions primarily written in Indonesian, though some contain mixed-language elements such as foreign term or regional dialect. The dataset is divided into 70% training data ad 30% testing data. Experimental results show that model achieved an accuracy of 90,83%, precision of 97,01%, recall of 87,84%, and f1-score of 92,90%. Evaluation of three NER rules (exact match keyword, prepositional patterns, and descriptive structures) indicates that the combination of all rules yields the highest f1-score (89%), while the best-performing individual rule is the prepositioning pattern (74%). These results demonstrated strong performance in processing varied and unstructured Instagram captions. The combinations of SVM and NER rule-based prove effective in identifying and classifying spatial information into two classes Contains Location and No Location. This approach shows potential for implementation in text-based spatial analysis systems, such as location-based recommendation systems, geographic mapping, and location-based decision support systems. AbstrakPerkembangan teknologi digital yang pesat secara signifikan berpengaruh meningkatkan produktivitas dan kemudahan akses informasi dalam kehidupan sehari-hari, salah satunya penggunaan media sosial yang semakin meluas. Instagram merupakan salah satu platform yang banyak digunakan, dimana teks dalam caption memiliki informasi terkait lokasi yang dapat dimanfaatkan untuk analisis spasial. Penelitian ini bertujuan untuk mengidentifikasi dan mengklasifikasikan entitas lokasi dalam caption Instagram menggunakan algoritma Support Vector Machine (SVM) dengan pendekatan Named Entity Recognition (NER) rule-based. Metode yang digunakan meliputi ekstraksi fitur berbasis linguistik dengan konteks spasial eksplisit, lebelisasi data, pelatihan model, serta evaluasi kinerja model menggunakan matriks klasifikasi: akurasi, presisi, recall dan f1-score. Dataset terdiri dari 400 caption umumnya berbahasa Indonesia, namun terdapat unsur bahasa campuran seperti istilah asing atau bahasa daerah. Fokus utama penelitian diarahkan pada pengolahan dan pemahaman teks berbahasa Indonesia. Dataset dibagi menjadi 70% data training dan 30% data testing. Hasil pengujian menunjukkan bahwa model mendapatkan akurasi sebesar 90,83%, presisi 97,01%, recall 87,84% dan f1-score 92,90%. Evaluasi terhadap tiga rule NER (exact match keyword, pola preposisi, dan struktur deskriptif) menunjukkan bahwa pengenalan entitas berdasarkan gabungan seluruh rule memberikan f1-score tertinggi (89%), sementara rule individual terbaik adalah pola preposisi (74%). Nilai ini menunjukkan kinerja yang cukup baik dalam pengolahan caption Instagram yang variatif dan tidak terstruktur. Kombinasi metode SVM dan NER rule-based terbukti efektif dalam mengidentifikasi dan mengklasifikasi informasi spasial dalam dua kelas Contain Location dan No Location. Pendekatan ini berpotensi diterapkan pada sistem analisis spasial berbasis teks, seperti sistem rekomendasi lokasi, pemetaan geografis, dan pendukung keputusan berbasis lokasi.