IOT routing optimization by using the K-Means clustering algorithm and whale optimization algorithm

Abstract

Natural disasters such as landslides, floods, floods, fires, volcanic eruptions and damage caused by these events are global problems that end up causing financial and life losses. This situation is aggravated by changes in the planet's climatic conditions and is observed mainly in urban regions. In these regions, the destruction of the ecosystem is more accentuated and ends up affecting the environment and modifying the local climate due to pollution and lack of planning. Therefore, this project brings 3 main contributions: (i) the use and evaluation of new standards and emerging technologies from the IoT together with WSN for the collection of data in natural environments and its distribution, (ii) the use of the collected data for the prediction of natural disasters using Machine Learning (ML) techniques, with a case study on the characteristics of rivers and rainfall in Iraq and turkey and, (iii) the proposal of an architecture IoT-based and ML-based fault-tolerant for the system, which would comprise prediction models for autonomously detecting natural disasters, even when part of the system was compromised, and generating alerts for the local population.