Dzikri Rahadian Fudholi

Traffic congestion has been a major problem in big cities, including Yogyakarta, with negative impacts including time, economic, and psychological losses. Based on data from the Yogyakarta Special Region Transportation Agency and Yogyakarta City Transportation Agency, analysis of congestion level data, and field observations, it was found that one of the main causes of congestion on the most congested roads in Yogyakarta City is vehicles parked on the side of the road. The proposed solution involves roadside parking detection and warning using surveillance cameras integrated with Artificial Intelligence (AI). The proposed system involves vehicle detection with pre-trained deep learning models, parking detection algorithms with Intersection over Union (IoU) tracking, and alerts that are forwarded to motorists as well as authorities such as the Transportation Department and local traffic police. The Yogyakarta CCTV dataset is used to test parking detection using various models, such as YOLOv5-medium, YOLOv5-large, YOLOv7-tiny, and Haar Cascade. The model evaluation shows that YOLOv5-large provides the highest accuracy of 86.1% with a processing speed of 5.5 Frames Per Second (FPS) to perform parking detection. With this proposed system, this research can contribute to solving congestion problems and improving traffic conditions in Yogyakarta City.

Music enhancement through deep learning methodologies presents an innovative approach to refining and augmenting classical compositions. Leveraging a comprehensive dataset of classical piano MIDI files, this study employs LSTM networks with attention mechanisms for music refinement. The model, trained on diverse compositions, demonstrates proficiency in capturing tempo nuances but faces challenges in replicating varied pitch patterns. Assessments by 28 individuals reveal positive reception, particularly in melody integration, scoring notably high at 8 out of 10. However, while praised for cohesion, bass lines received slightly lower scores, suggesting opportunities for enhancing originality and impact. These findings underscore the LSTM model's capability to generate harmonious melodies and highlight refinement areas, particularly in innovating bass lines within classical compositions. This study contributes to advancing automated music refinement, guiding further developments in LSTM-based music generation techniques.

Earthquakes present significant risks to both human safety and infrastructure, emphasizing the need for precise prediction models to minimize their adverse effects. This study seeks to tackle the challenge of accurately forecasting the occurrence time of earthquakes by utilizing the LANL Earthquake dataset, which comprises seismic signals from a laboratory model emulating tectonic faults. In this study, we employed the ARIMA model and compared it with Linear Regression to predict earthquake occurrences. Our findings demonstrate that the ARIMA (1,1,1) model surpasses other models, achieving the lowest MAE of 0.110628. The validity of the model's assumptions is confirmed through the Ljung-Box and Jarque-Bera tests, which verify the absence of autocorrelation and the normal distribution of residuals, respectively.