Nowadays machine learning is everywhere, and every day we discover new ways to implement it. One usage that is growing very fast is building price estimation systems. Many platforms are opting to implement this kind of technology to offer their users a better experience when visiting their platform. In this document, you will find a development and analysis of a price recommendation system based on machine learning algorithms. In the society that we live today people have many objects that they just do not use anymore, but because of this fact, many companies have created platforms for selling used and new items. When a company develops its own product they can know how much it really costs, based on their experience building it. But when it comes to selling used products you can miss the price when you are using only your intuition and in some cases even emotions attached to the item. In this document, we will develop a solution that fulfills this need. You will be able to see how text can be processed to build information that a computer can work with, this is known as Natural Language Processing (NLP). We will also create a KNN (k-nearest neighbors) which is going to be the tool that we will use to predict the output price of the item. I will guide through the whole process to build this kind of system, from analyzing the data to predicting output prices. It will be also shown the precision and the score of the algorithm so that we can see how good or how bad different solutions can work