Susetyo, Yosia Adi
Unknown Affiliation

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

LSTM-IOT (LSTM-based IoT) untuk Mengatasi Kehilangan Data Akibat Kegagalan Koneksi Susetyo, Yosia Adi; Parhusip, Hanna Arini; Trihandaru, Suryasatriya; Susanto, Bambang
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 12 No 1: Februari 2025
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2025129157

Abstract

Masalah dalam industri terkait kehilangan data suhu dan kelembaban sering terjadi akibat gangguan perangkat atau hilangnya koneksi. Data ini penting untuk menentukan kelayakan produk yang akan didistribusikan. Untuk mengatasi permasalahan tersebut, dikembangkan inovasi LSTM-IOT, yaitu perangkat IoT yang terintegrasi dengan model Long Short-Term Memory (LSTM) dalam arsitektur Environment Intelligence. Arsitektur ini telah dioptimalkan melalui eksperimen menggunakan berbagai jenis optimizer, seperti Adam, RMSprop, AdaGrad, SGD, Nadam, dan Adadelta. Dari hasil optimasi, kombinasi Nadam Optimizer dengan arsitektur terpilih menunjukkan kinerja unggul dengan nilai Mean Square Error (MSE) sebesar 5,844 x10⁻⁵, Mean Absolute Error (MAE) sebesar 0,005971, dan Root Mean Square Error (RMSE) sebesar 0, 007645. Arsitektur Environment Intelligence versi (a) dengan Nadam Optimizer terbukti paling efektif dalam memproses data sensor, sehingga dipilih untuk integrasi dengan perangkat LSTM-IOT. Implementasi LSTM-IOT dalam skenario dunia nyata dilakukan pada wadah web lokal yang memungkinkan akses real-time ke data suhu dan kelembaban di berbagai lokasi. Halaman web berbasis Streamlit ini menampilkan visualisasi data, performa LSTM, dan hasil prediksi. Uji fungsional menunjukkan bahwa LSTM-IOT memenuhi kebutuhan perusahaan, termasuk penyimpanan data dalam database internal serta prediksi kondisi lingkungan hingga 150 menit ke depan. Dengan fitur prediksi dan pemantauan yang canggih, perangkat ini memberikan solusi efisien dan bernilai tinggi bagi perusahaan dalam memantau kondisi lingkungan secara akurat dan proaktif.   Abstract Problems in the industry related to temperature and humidity data loss are often caused by device interference or loss of connection. This data is important to determine the feasibility of the product to be distributed. To overcome these problems, an LSTM-IOT innovation was developed, namely an IoT device that is integrated with the Long Short-Term Memory (LSTM) model in the Environment Intelligence architecture. This architecture has been optimized through experiments using different types of optimizers, such as Adam, RMSprop, AdaGrad, SGD, Nadam, and Adadelta. From the optimization results, the combination of Nadam Optimizer with the selected architecture shows superior performance with a mean square error (MSE) value of 5.844 x 10⁻⁵, a mean absolute error (MAE) of 0.005971, and a root mean square error (RMSE) of 0.007645. The Environment Intelligence architecture version (a) with Nadam Optimizer proved to be the most effective in processing sensor data, so it was chosen for integration with LSTM-IOT devices. The implementation of LSTM-IOT in real-world scenarios is carried out on a local web container that allows real-time access to temperature and humidity data in various locations. This Streamlit-based webpage displays data visualizations, LSTM performance, and prediction results. Functional tests show that LSTM-IOT meets the needs of the company, including data storage in an internal database and prediction of environmental conditions for up to the next 150 minutes. With advanced prediction and monitoring features, these devices provide efficient and high-value solutions for companies to monitor environmental conditions accurately and proactively.
Development of Staff Evaluation Software Based on Association Matrix Methods and Data Mining Using the Streamlit Framework Susetyo, Yosia Adi; Parhusip, Hanna Arini; Trihandaru, Suryasatriya
JUITA: Jurnal Informatika JUITA Vol. 12 No. 2, November 2024
Publisher : Department of Informatics Engineering, Universitas Muhammadiyah Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30595/juita.v12i2.23300

Abstract

This study discusses evaluating employee performance in microbiology laboratories using an association matrix implemented in web-based software with the Streamlit framework. The purpose of the research is to improve the employee performance evaluation process, which previously used conventional methods. This software is built from a sample receipt recording history data stored in a MySQL database. The initially unstructured data was processed using Python libraries such as NumPy, Matplotlib, Pandas, and Difflib to generate personnel evaluation information such as specialization, task duration, workload, and individual competencies. This software can provide a fast and accurate performance assessment according to the evaluation period. In a test with the System Usability Scale (SUS), the software scored 75.83, which was rated "good.". These results show that the software is easy to use and can improve the efficiency of employee performance evaluation. Follow-up tests with questionnaires given to 18 users showed that this system was preferable to previous conventional methods. This software helps laboratory managers evaluate employee performance effectively and efficiently.
Human Capital Decision Intelligence (HCDI) architecture in microbiology laboratory based on machine learning and operations research models Trihandaru, Suryasatriya; Susetyo, Yosia Adi; Parhusip, Hanna Arini; Susanto, Bambang
International Journal of Advances in Intelligent Informatics Vol 11, No 4 (2025): November 2025
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v11i4.1676

Abstract

The Human Capital Decision Intelligence (HDCI) system integrates human-computer interaction in a microbiology laboratory that uses machine learning and operational research to classify new tasks and then recommend assignments to each person. The models evaluated in building this system are Support Vector Machine, Gaussian Naive Bayes, Multinomial Logistic Regression, and Artificial Neural Network. The results of the research show that the ANN model is the most consistent and reliable across various training ratios, as indicated by the model's goodness parameters. The selected ANN model is combined with a linear programming approach to optimize workload distribution. The integrated system successfully manages new job scenarios and recommends staff based on competencies and availability. It also ensures assignments do not exceed maximum workload limits and finds alternatives when key staff are unavailable. The implementation of the HDCI system has a positive impact on various factors, including the fair distribution of tasks, enhanced staff performance monitoring, and significantly improved operational efficiency and human resource management in the microbiology laboratory. The system is designed to be easy to use and support collaboration between laboratory staff and computational models. The system is not only advanced in supporting personnel management decision-making, but it can also demonstrate how artificial intelligence and operations research systems can be combined to address the needs of the microbiology laboratory environment.