Contextual Awareness System for Landslide Risk Recommendation in Crypto-Spatial

Landslides result in significant damage to the impacted regions. The precision of predictions and welldefined boundaries of geospatial data make landslide risk assessment essential. Unfortunately, integrated environmental impacts are often not even correctly observed or considered in traditional methods, leading to few recommendations. This study introduces a contextual awareness system for landslide risk recommendations in crypto-spatial. It employs XGBoost to predict landslide risks by utilizing environmental data sourced from Google Earth Engine, along with crypto-spatial technology, to improve the integrity and transparency of geospatial information. The experimental findings reveal that the model achieves an accuracy rate of 95%, alongside a precision of 93% and a recall of 92%, in addition to an F1score of 92.5%. This superior performance surpasses that of traditional methods such as Random Forest and Support Vector Machines (SVM). This study underlines a notable advancement in leveraging machine learning techniques for disaster risk assessment, supported by contextual awareness and crypto-spatial integration. This study aims to propose a controlled intelligent recommendation system so that it can reduce the risk of landslides and help to make better decisions based on the available data.