The cost of future claims is an important study to determine the pure rate of any coverage offered by an insurer. To do this, the behavior of claims must be analyzed and a specific trend that can be applied to the cost of insurance must be determined. For this reason, the opportunity arises to use predictive models present in Machine Learning that help optimize processes and have a broader notion of claims. On this occasion, various Machine Learning models are applied to an artificial database whose behavior is based on real claims associated with work accidents. This base is divided into training and validation samples, in order to train the suggested models and validate the prediction capacity by applying measures to different data manipulation scenarios.
