An enhanced discovery of multiple natural disasters using machine learning model

Today, many natural disasters occur in the world. Damage caused during disaster assessment has been a vital process. Traditional methods of assessing the cause of damage are neither fast nor efficient. In disaster error estimation, progress has been made in recent years. Internet resources serve as a dynamic facilitator to obtain data for the model and produce the desired output. The proposed hybrid CNN helps in detecting and identifying the natural disaster damage and also precisely assesses the damages with higher accuracy. Hybrid CNN performance metrics, such as accuracy, precision, recall, and F1 score, are compared with logistic regression, support vector machine, gradient boosting, and random forest algorithms in the processing of imagery data with natural disaster regions and provide the cost estimation with any objects precisely over the higher accuracy rate of greater than 96%.