Claim Missing Document
Check
Articles

Inovasi AI: Sosialiasi Robot Personal Assistant untuk Digitalisasi KIR 03 Lindawati, Lindawati; Salamah, Irma; Valerie, Michelle; Kusumanto, RD
Jurdimas (Jurnal Pengabdian Kepada Masyarakat) Royal Vol. 6 No. 4 (2023): Oktober 2023
Publisher : STMIK Royal

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33330/jurdimas.v6i4.2600

Abstract

Penerapan teknologi kecerdasan buatan (AI) terus mengalami kemajuan pesat dan memiliki potensi yang luar biasa dalam mendukung transformasi digital di berbagai sektor. Dalam rangka pengembangan ekstrakurikuler di KIR 03, dilaksanakan sosialisasi inovasi AI dengan memanfaatkan robot personal assistant guna mendorong digitalisasi. Tujuan utama dari kegiatan pengabdian ini adalah untuk meningkatkan pemahaman peserta mengenai manfaat dan potensi penggunaan teknologi AI dalam konteks kegiatan ekstrakurikuler. Pendekatan yang digunakan meliputi penyampaian materi presentasi yang mendalam, interaksi tanya jawab yang aktif, serta dokumentasi kegiatan secara komprehensif. Hasil analisis data dari kuesioner menunjukkan peningkatan signifikan dalam pemahaman peserta setelah dilakukan sosialisasi. Respon positif peserta terhadap manfaat dan kemudahan penggunaan robot personal assistant menggambarkan kesuksesan sosialisasi ini. Sosialisasi ini memberikan implikasi potensial yang kuat dalam mendukung digitalisasi di KIR 03 dan berkontribusi bagi masyarakat pada umumnya.
Sistem Deteksi Gas Pintar Berbasis IoT dan Terintegrasi Fuzzy-Logic untuk Keamanan Distribusi Gas secara Realtime Azizah, Putri Nur; Taqwa, Ahmad; Salamah, Irma
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7331

Abstract

Abstract−LPG gas is a widely used fuel for daily needs in households, industry, and commercial sectors. Although easy to use and affordable, LPG contains highly flammable compounds that can cause fires and explosions, especially if leaks go undetected. Field surveys show that most gas agents or depots still use manual methods relying on the sense of smell to detect gas leaks. This approach does not provide optimal or accurate results, making it ineffective and potentially harmful to health when excessive gas is inhaled. Therefore, this research aims to design a gas leak detection system based on the Internet of Things (IoT) using the Fuzzy Tsukamoto algorithm integrated with the Blynk application. The method involves the design of hardware and software using three sensors as input parameters: MQ-6 (gas), DHT22 (temperature), and Flame Sensor (fire), which are processed by the ESP32 microcontroller through fuzzy logic rules. The system outputs include a visual LED indicator, buzzer activation, status display on the LCD, notifications via Blynk, and automatic fan response to neutralize the gas. Based on results simulation and testing under three environmental condition scenarios, the system is able to detect gas leaks with average error of 0.315% and accuracy of 90.55%. This study demonstrates a reliable, effective, and responsive gas leak detection system. It is expected that the system can minimize the potential dangers of gas leaks and enhance gas storage safety.
Knowledge Sharing Behavior Model of Polytechnic Lecturer Kusumanto, RD; Salamah, Irma
International Journal of Science, Technology & Management Vol. 4 No. 1 (2023): January 2023
Publisher : Publisher Cv. Inara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46729/ijstm.v4i1.696

Abstract

Knowledge sharing is very important for lecturers in implementing the tri dharma of higher education. By sharing knowledge, it will create creative ideas and innovations for lecturers. This study aims to find out how the behavior of sharing knowledge of polytechnic lecturers. Research on the knowledge sharing behavior of polytechnic lecturers has never been carried out. The research sample was 200 lecturers at the first six polytechnics in Indonesia. The behavior of sharing knowledge is seen from three factors: knowledge donating, knowledge collecting, and technology. The data was processed using LISREL 8.80. The results of the research are the dominant dimension in supporting the sharing of lecturers' knowledge. Polytechnics lecturer also tend to collecting more knowledge than share knowledge. This can be due to the lack of motivation of lecturers to share and competition. For this reason, polytechnic university management needs efforts to improve the culture of knowledge sharing among lecturers.
Sistem Mobile-Cloud untuk Identifikasi Jenis Kayu, Estimasi Umur, dan Analisis Performansi Transmisi Citra Menggunakan CNN dan Random Forest Abdillah, Depran; Salamah, Irma; Ciksadan
JEPIN (Jurnal Edukasi dan Penelitian Informatika) Vol. 12 No. 2 (2026): Volume 12 No 2
Publisher : Program Studi Informatika

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

Abstract

Identifikasi jenis kayu pada industri mebel masih banyak dilakukan secara manual, sehingga berpotensi menimbulkan ketidaktepatan dalam penentuan jenis dan estimasi umur kayu. Penelitian ini mengembangkan aplikasi mobile-cloud berbasis Android untuk klasifikasi jenis kayu dan estimasi umur kayu pada Mebel Karya Bersama Palembang. Model Convolutional Neural Network (CNN) digunakan untuk mengklasifikasikan lima jenis kayu, yaitu Jati, Meranti, Seru, Tembesu, dan Unglen, berdasarkan citra permukaan kayu, sedangkan Random Forest digunakan untuk mengestimasi kategori umur kayu berdasarkan fitur hasil ekstraksi CNN. Kinerja model CNN yang dievaluasi menunjukkan pencapaian akurasi pada angka 91%, sedangkan model Random Forest memperoleh akurasi 98,4%. Sistem yang dibangun diimplementasikan dalam arsitektur mobile-cloud dan mampu menampilkan hasil prediksi jenis kayu, estimasi umur, serta performansi transmisi citra berdasarkan parameter delay, throughput, dan packet loss pada jaringan WiFi dan 4G LTE. Penelitian ini menunjukkan bahwa aplikasi yang dikembangkan dapat mendukung identifikasi kayu secara lebih cepat, praktis, dan terkomputerisasi.