Writer : Mochammad Agus Yahya, Zefanya Soplantila Study Program : Computer Science Title : Development of a Real-Time Soil Moisture Monitoring System with Integration of Streaming Data and Machine Learning Models to Support Prefailure Inspection on Slope Stability Counselor : Made Harta Dwijaksara, Ph.D., Abdul Halim Hamdany, Ph.D. The rapid expansion of urbanization and infrastructure growth in Indonesia has significantly altered the physical landscape, affecting slope stability and increasing the risk of landslides. Additionally, increased water pressure in soil, particularly in unsaturated soil areas, can lead to landslides. Therefore, knowledge of soil moisture in an area is crucial for understanding slope stability. This study aims to develop a real-time soil moisture monitoring system based on IoT, integrating a data streaming system and implementing machine learning models to anticipate and mitigate landslide risks. The development of the system is divided into two main parts: the development of a data streaming system infrastructure to integrate data flows from various IoT devices, and the implementation of machine learning models to predict soil moisture and soil suction. The data streaming system allows sensor data to be stored in a database and visualized through a user-friendly web interface in real-time, while the machine learning models can estimate soil conditions in areas not monitored by sensors to effectively conduct pre-failure inspections. Test results show that the system can distribute data from 500 sensors with a latency of less than 1 second, demonstrating good scalability with low and stable CPU usage, ensuring quick and efficient soil condition monitoring. Furthermore, the system can also perform predictions using both conventional and traditional machine learning algorithms that utilize several weather features or solely rely on spatial interpolation, although accuracy is not the main focus of this study.