The ability to know how we communicate and interact with the people around us is fun- damental in the times we live. Thanks to all the technologies we have at our disposal on a daily basis, we are able to meet a wide variety of people from all over the world, which allows us to know how they live, what culture they have, and how they speak and relate to the world. As we change as a society and new habits emerge, our way of speaking changes, leading to the creation of new words with completely new meanings which are introduced into our language. These new words are called neologisms. As mentioned, knowing these new words is essential, so ways of detecting neologisms have emerged to be at the forefront of this sector. The way to detect them is mainly based on the use of natural language processing techniques. However, a new approach has emerged in which neologisms are detected by their usage trend. This project consists of detecting neologisms using the latter technique, which consists of being able to adjust the popularity data of a neologism according to an epidemiological model called SIR. This model is used because it is used to mathematically express the spread of an infection, which starts with very few infected people, but as time goes by, the number of infected people increases, until finally they end up recovering and the infection ends. The same can be taken to the field of neologisms, which very few people use at first, but over time they grow in popularity, until at a certain point people stop using them