Currently there is a large amount of information circulating through social networks, this does not always tend to be true and in the case of natural disasters its falsity could have quite consequences such as mass hysteria in the population. To avoid this, an efficient analysis was proposed to check tweets with false information using natural language processing algorithms.