Claim Missing Document
Check
Articles

Pengenalan Gerakan Olahraga Berbasis (Long Short- Term Memory) Menggunakan Mediapipe Daniel Tanugraha, Filbert; Pratikno, Heri; Musayyanah; Indah Kusumawati, Weny
Journal of Advances in Information and Industrial Technology Vol. 4 No. 1 (2022): May
Publisher : LPPM Telkom University Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52435/jaiit.v4i1.182

Abstract

Olahraga Yoga menjadi gaya hidup yang sangat trend di kalangan masyarakat. Untuk memastikan gerakan Yoga yang dilakukan benar atau tidak, maka perlu adanya sistem deteksi/pengenalan dan klasifikas yang terkenal dengan istilah HAR (Human Activity Recognition). Sistem pengenalan gerakan olahraga aktivitas manusia, seperti Olahraga Yoga yang dikenal dengan Tree Pose, T-Pose, Warrior II Pose. Gerakan ini dapat dideteksi secara real time menggunakan MediaPipe agar gerakan tersebut dapat dikenali oleh komputer dengan akurat. Sedangkan untuk klasifikasi olahraganya manusia menggunakan Long Short-Term Memory (LSTM) dengan memanfaatkan keypoints dari urutan video tersebut. Pengujian klasifikasi dan deteksi gerakan dilakukan oleh beberapa sampel dengan lokasi yang berbeda menggunakan kamera handphone dalam pengambilan video menggunakan aplikasi DroidCam dalam pendeteksian keypoint. Pada penelitian ini, hasil training didapatkan akurasi sebesar 91% dan loss sebesar 29%. Pengenalan gerakan T-Pose, Warrior II Pose, dan Tree Pose berturut-turut adalah 100%, 85% dan 80%.
Monitoring Dan Controlling Pemberi Pakan Ikan Otomatis Pada Akuarium Menggunakan MQTT Atthaariq Maulana, Muhammad; Harianto; Musayyanah; Indah Kusumawati, Weny
Journal of Advances in Information and Industrial Technology Vol. 4 No. 2 (2022): November
Publisher : LPPM Telkom University Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52435/jaiit.v4i2.248

Abstract

Budidaya ikan hias pada akuarium berkembang pesat. Memberi makan ikan dengan waktu yang tidak teratur dapat mengakibatkan kematian. Awalnya pakan diberikan secara manual, namun saat ini tercipta alat pemberi pakan ikan secara otomatis. Pemberi pakan ikan yang dibuat secara otomatis ini, memakai komunikasi MQTT supaya bisa diakses dan dikontrol dari jarak jauh menggunakan koneksi internet. Peralatan ini mempunyai beberapa fungsi yaitu: pemberian pakan secara otomatis ataupun manual, monitoring (pemantauan) suhu air, pemantauan status pompa air, kontrol lampu akuarium otomatis, dan otomatisasi heater. Monitoring dan kontroling menggunakan aplikasi MQTT Dashboard. Sensor DS18B20 mempunyai fungsi sebagai pendeteksi suhu air. Dari hasil pengujian, hasil akurasi error sensor ini sebesar 1.1%. Ukuran pakan ikan yang dapat dikeluarkan dan ditampung adalah 0.8 mm sampai dengan ukuran 1 mm. Keseluruhan pengujian komunikasi MQTT memperoleh akurasi 100%.
Prototipe Sistem Monitoring Suhu, Ketinggian Air, dan Kontrol Otomatis pada Budikdamber Berbasis IoT Yazid, Abdul; Weny Indah Kusumawati; Febriliana, Ridha
Journal of Advances in Information and Industrial Technology Vol. 5 No. 2 (2023): Nov
Publisher : LPPM Telkom University Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52435/jaiit.v5i2.371

Abstract

Penelitian bertujuan untuk memanfaatkan lahan terbatas yang tersedia di perkotaan dengan sebaik-baiknya dengan membuat sistem kendali dan sistem pemantauan pembudidayaan ikan dalam ember yang didasarkan pada Internet of Things. Pengunaan Internet of Things yang cerdas dan sensor-sensor pada penelitian ini memungkinkan segala pemantauan dan pengaturan keadaan budidaya ikan dalam ember secara remote. Untuk meningkatkan keberhasilan budidaya ikan dan meningkatkan kesejahteraan ikan yang dipeliharanya, teknologi ini memungkinkan pembudidaya ikan untuk memantau dan mengendalikan budidaya secara lebih efektif dan cepat, sehingga sistem ini memiliki potensi untuk secara positif mempengaruhi perluasan budidaya ikan ember dalam konteks perkotaan yang terbatas. Hasil pengujian Sensor DS18B20 menunjukkan bahwa terdapat nilai kesalahan (error) sebesar 1.6%. Hasil pengujian jarak Sensor Ultrasonik menunjukkan bahwa terdapat nilai kesalahan sebesar 2.5%. Pada uji coba komponen penggerak (servo) memiliki tingkat keberhasilan 93.3%. Pada pengujian modul relay memiliki tingkat keberhasilan adalah 100%. Pada pengujian keseluruhan sistem didapatkan tingkat keberhasilan sebesar 100%.
Klasifikasi Emosional Ulasan Pelanggan dengan Pendekatan NLP menggunakan Metode Ensemble dan ROS Noorizki, Adisaputra Zidha; Pratikno, Heri; Kusumawati, Weny Indah
Techno.Com Vol. 23 No. 4 (2024): November 2024
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/tc.v23i4.11559

Abstract

Konsep Orientasi Pelanggan sangat penting bagi perusahaan untuk berkembang di era saat ini, dengan memanfaatkan teknologi untuk mendapatkan wawasan yang mendalam tentang perilaku pelanggan mereka. Salah satu alat teknologi tersebut adalah pembelajaran mesin, khususnya yang menggunakan pendekatan pemrosesan bahasa alami (NLP). Penelitian ini menggunakan lima algoritma yang berbeda dan menggabungkan berbagai metode untuk meningkatkan kinerja model machine learning. Melalui penerapan teknik-teknik seperti random over-sampling (ROS) dan ensemble learning, akurasi prediksi keseluruhan untuk kelas minoritas meningkat secara signifikan. Model ensemble yang diintegrasikan dengan ROS mencapai akurasi 0,90 dan mean square error 0,91, mengungguli algoritma lain yang diuji dalam penelitian ini. Pendekatan yang dioptimalkan ini tidak hanya menunjukkan keefektifan pemanfaatan teknologi untuk sebuah perusahaan dapat menerapkan strategi yang berpusat pada pelanggan, tetapi juga menyoroti pentingnya peningkatan metodologi dalam pemodelan prediktif untuk keberlanjutan bisnis.   Kata kunci: Klasifikasi Emosi, Pembelajaran Mesin, Pemrosesan Bahasa Alami, Hard Voting, Random Over Sampling.
Ensemble Voting Method for Phonocardiogram Heart Signal Classification Using FFT Features Noorizki, Adisaputra Zidha; Pratikno, Heri; Kusumawati, Weny Indah
Journal of Applied Informatics and Computing Vol. 8 No. 2 (2024): December 2024
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v8i2.8704

Abstract

Heart disease is still one of the leading causes of death worldwide, hence the need for effective diagnostic tools. Phonocardiogram (PCG) signals have been explored as a complementary approach to electrocardiogram (ECG) to detect cardiac abnormalities. This research investigates the classification of PCG signals using Fast Fourier Transform (FFT) features and deep learning models, including Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Temporal Convolutional Network (TCN). Hyperparameter tuning, particularly learning rate adjustment, is applied to optimize the performance of the models. The results show that the GRU and TCN models outperform the LSTM, achieving up to 92% accuracy at a learning rate of 0.0001. Ensemble learning with soft voting was also applied to combine the strengths of each model. Although the ensemble model showed strong performance with 92% accuracy and ROC AUC of 0.9636, it did not provide significant improvement over the base model. This finding highlights the importance of hyperparameter tuning in model optimization, with GRU and TCN showing slightly better performance in the time series classification task. This study concludes that ensemble learning offers stability but does not significantly improve classification accuracy beyond a well-tuned base model.
Lighter as flame control and temperature control in milk pasteurization system Febriansyah, Rizky Alief; Kusumawati, Weny Indah; Harianto; Pauladie Susanto
Matrix : Jurnal Manajemen Teknologi dan Informatika Vol. 13 No. 3 (2023): Matrix: Jurnal Manajemen Teknologi dan Informatika
Publisher : Unit Publikasi Ilmiah, P3M, Politeknik Negeri Bali

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31940/matrix.v13i3.115-129

Abstract

Milk pasteurization involves heating milk at a specific temperature below its boiling point as a method to keep the resulting dairy product retaining the shape and taste characteristics of fresh milk. In several experiments, the use of fuzzy control systems has been tested to regulate the temperature in the milk pasteurization process and to time the ignition of the stove flame. However, this fuzzy approach still causes the system to give an unstable response and irregular stove flame. In order to study milk pasteurization, the Low Temperature Long Time (LTLT) method is used which is implemented automatically through a PID control system. This method serves to maintain nutritional quality by keeping the pasteurization temperature at a setpoint of 62°C. This control involves the use of servo actuators and electric lighters that are automatically regulated with the help of flame sensors. The flame sensor detects the presence of flame and ensures that the flame remains lit throughout the pasteurization process. At the end of the process, the flame sensor plays a role in breaking the flame by setting a certain threshold. This sensor operates within 10 CM of the flame source and will produce an analog output with a maximum value of 4000 when the flame is lit. When the flame is extinguished, the analog output of the sensor will reach a value of 4095. Testing of the milk pasteurization automation system is given a value for each PID resulting from the value of Kp = 31.8, Ki = 115.6, Kd = 4.4 and obtained a rise time value of 0.39 minutes, 0.61%, settling min of 60.88, and settling max of 62.38.
Monitoring and temperature control system for fish farming in an IOT-based bucket using an android application Yazid, Abdul; Kusumawati, Weny Indah; Harianto; Susanto, Pauladie
Matrix : Jurnal Manajemen Teknologi dan Informatika Vol. 14 No. 2 (2024): Matrix: Jurnal Manajemen Teknologi dan Informatika
Publisher : Unit Publikasi Ilmiah, P3M, Politeknik Negeri Bali

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31940/matrix.v14i2.50-65

Abstract

This study seeks to create and implement an Internet of Things (IoT)-based system for monitoring and controlling temperature in fish farming within a bucket (Budikdamber) using temperature sensors. The main focus is to set the optimal temperature range for catfish farming in the Budikdamber, ensuring environmental conditions that suit the needs of catfish. Automatic controls are integrated to keep the water temperature in the bucket within the desired range. The system is also capable of automatically activating or deactivating devices such as the Peltier, lights, filling pump, and draining pump to achieve the appropriate temperature. In the test using a 50-litre bucket, a thorough evaluation and testing showed that the DS18B20 temperature sensor had an error rate of 0.8826%, while the Ultrasonic sensor had an error rate of 0.737%. Nonetheless, the other components in the device operate optimally with performance levels reaching 100%, according to their functions. Testing of the Android App on this device achieved an accuracy rate of 100%, indicating that this device can be effectively monitored and controlled remotely through the Android App. The conclusion of this research provides insight into the performance of temperature sensors, system optimization, and the effectiveness of remote control using Android Apps in the context of fish farming.  
Pelatihan Meningkatkan Kompetensi Guru MGMP TIK Surabaya: Menguasai Arduino untuk Pendidikan Kreatif Kusumawati, Weny Indah; Harianto; Heri Pratikno
Ekobis Abdimas Vol 6 No 1 (2025): Juni 2025
Publisher : Fakultas Ekonomi, Universitas PGRI Adi Buana Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36456/ekobisabdimas.5.1.10103

Abstract

Information and Communication Technology (ICT) is a field that teaches technology skills and how to use information systems in many areas of life. In schools, ICT is taught so that students can recognize and have the ability to use technology in a good, efficient and ethical way. MGMP TIK Surabaya is a community of teachers who teach subjects related to Information and Communication Technology in Surabaya. MGMP, which means Musyawarah Guru Mata Pelajaran, aims to improve the professional and teaching abilities of ICT teachers, strengthen cooperation among teachers, and develop teaching methods and materials that are in line with technological advances. The goal of this Community Service program is to provide technical skills training as well as improve critical and creative thinking abilities in students. Through project-based methods, Arduino offers a practical and immersive learning experience, which prepares students to face the challenges of an increasingly modern and technological world. The problems faced in Arduino education can be solved with appropriate training and collaborative and interactive teaching methods. The implementation of community service activities has been completed in accordance with the specified time. This community service activity is carried out through the delivery of materials, discussions, and practices carried out in a hybrid manner. The success of this activity can be seen from how easy it is for MGMP ICT teachers to use the application and Arduino module package, and all teachers hope that there will be another community service program. The results of the questionnaire survey showed that the teachers mostly gave “strongly agree” and “agree” answers to all the questions asked. 
Webinar Perkembangan AI dan Robotika di Industri Kesehatan Puspasari, Ira; Kusumawati, Weny Indah
Ekobis Abdimas Vol 6 No 1 (2025): Juni 2025
Publisher : Fakultas Ekonomi, Universitas PGRI Adi Buana Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36456/ekobisabdimas.5.1.10346

Abstract

The development of artificial intelligence (AI) and robotics has revolutionized various aspects of modern healthcare, including diagnosis, surgery, and rehabilitation. The webinar, titled "Developments in AI and Robotics in the Health Industry," was organized by the Dinamika University Computer Engineering Undergraduate Study Program as a form of community service to increase technological literacy in the health sector. This activity aimed to introduce AI technology and medical robotics and explain the application and ethical, technical, and policy challenges in its implementation. The webinar, which was held online via Zoom, was attended by 160 participants from various backgrounds, such as students, lecturers, health practitioners, and technology developers. The material covered the evolution of robotic systems, the level of autonomy of medical robots, and the potential integration of AI for clinical decision-making. Survey results showed that participants were very enthusiastic and highly rated the relevance of the theme, the competence of the speakers, and the usefulness of the material. This positive feedback from the participants reassures the audience about the quality and relevance of the webinar's content, instilling confidence in the potential of AI and robotics in healthcare. The webinar also had a significant impact on interdisciplinary research, inspiring new collaborations and open discourse on academic cooperation. The webinar also encouraged the utilization of AI and robotics technology in the Indonesian health system, contributing to the digital transformation of the health sector and expanding public understanding of the latest medical technology innovations.
Penerapan Teknologi “KOPKA” (Kran Otomatis dan Pengukur Kapasitas Air) pada Bak Penampung Air Drip Irrigation untuk Mempermudah Kerja Kelompok Tani Banyu Urip Desa Tanggumong Kecamatan Sampang Weny Indah Kusumawati; Stephanie Astrid Ayu; Musayyanah, Musayyanah; Yunanto Tri Laksono; Charisma Dimas Affandi
PaKMas: Jurnal Pengabdian Kepada Masyarakat Vol 4 No 2 (2024): November 2024
Publisher : Yayasan Pendidikan Penelitian Pengabdian Algero

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54259/pakmas.v4i2.3307

Abstract

The Banyu Urip Farmer Group faces challenges in managing water for agricultural irrigation purposes. One of the main problems is that irrigation in the fields used still uses manual watering techniques, so it takes a long time to do watering. To overcome this problem, the Banyu Urip Farmer Group collaborated with Universitas Dinamika to install an Automatic Faucet and Water Capacity Meter system. This Automatic Faucet and Water Capacity Meter system is an innovation that helps improve irrigation efficiency and water management on farms. This technology provides convenience for farmers in terms of water control, which has a positive impact on resource use efficiency, and supports sustainable agriculture in Tanggumong Village, Sampang District. Community service activities have been completed according to the predetermined schedule. The success of this activity can be measured by the ease felt by Poktan in operating the application to monitor equipment, being able to fill in content on Instagram, and their enthusiasm to take part in the next community service programs.