Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Jurnal Teknik Informatika (JUTIF)

POLAK-RIBIERE CONJUGATE GRADIENT ALGORITHM IN PREDICTING THE PERCENTAGE OF OPEN UNEMPLOYMENT IN NORTH SUMATRA PROVINCE Amalya, Nanda; Solikhun, Solikhun
Jurnal Teknik Informatika (Jutif) Vol. 5 No. 1 (2024): JUTIF Volume 5, Number 1, February 2024
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2024.5.1.1047

Abstract

The economic problem that has a direct impact on human life and welfare is unemployment. One of the cities in Indonesia with the highest unemployment rate is North Sumatra Province. For example, Tebing Tinggi City had the highest unemployment rate of 9.73% in 2017, while Nias Selatan had the lowest percentage of 0.31%. This research is important to do in order to anticipate the unemployment rate in North Sumatra for any party, be it the government or the private sector, so that they can work together to overcome the problem of unemployment in the future which is the main problem in the economy. For example, the government creates programs to help reduce the number of unemployed, provide preparation or do other things, helping people to become more imaginative and have skills so they can compete in the world market. Predicting unemployment has been the subject of many studies, one of which is by utilizing artificial neural networks. This study aims to predict the percentage of unemployed in North Sumatra from 2022 to 2026, using the Backpropagation Neural Network Algorithm, the Conjugate Gradient Polak-Ribiere method and Matlab version 2011 for research and data analysis. This research utilizes open action rate stimulation data for the population of North Sumatra based on residents aged over 15 years from 2017 to 2021. Using five architectural models, namely: 4-50-1, 4-55-1, 4-70- 1, 4-75-1, and 4-77-1. The final results were obtained using the most accurate architectural model, namely model 4-75-1 which has a Mean Squared Error (MSE) of 0.0000004288 and an accuracy rate of 100% with a time of 00.09 at epoch 452.
Deep Learning Based MobileNet Optimization For High Accuracy Classification Of Toddler Stunting Wibowo, Anan; Sembiring, Rahmat Widia; Solikhun, Solikhun
Jurnal Teknik Informatika (Jutif) Vol. 7 No. 2 (2026): JUTIF Volume 7, Number 2, April 2026
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2026.7.2.5382

Abstract

This study aims to develop and optimize a MobileNet-based deep learning model for toddler stunting classification using whole-body images. A progressive optimization strategy was applied through three scenarios: (1) a baseline MobileNet feature-extraction model, (2) an optimized fine-tuned model, and (3) a final model enhanced with an adaptive ReduceLROnPlateau scheduler. Using a private dataset of 571 images, the proposed model achieved significant improvements—from 97.47% accuracy in the baseline model to a perfect 100% accuracy, precision, recall, and F1-score in the final scenario. These results highlight the novelty of this study, namely the use of whole-body images combined with progressive MobileNet optimization, which substantially outperforms prior studies relying solely on facial image analysis. The proposed approach demonstrates strong potential as a highly accurate and efficient computational tool for clinical stunting screening.