The talk is aimed at disseminating the work done over the past 2 years in constructing a data set for astronomical transient events (objects in the sky with aperiodic behaviour...more in the talk :D) and classifying them using state of the art machine learning methods. The abstract of the paper that summarizes our work can be found below: We introduce MANTRA, an annotated dataset of 4869 transient and 16940 non-transient object lightcurves built from the Catalina Real Time Transient Survey. We provide public access to this dataset as a plain text file to facilitate standardized quantitative comparison of astronomical transient event recognition algorithms. Some of the classes included in the dataset are: supernovae, cataclismic variables, active galactic nuclei, high proper motion stars, blazars and flares. As a complement to the dataset, we experiment with multiple data pre-processing methods, feature selection techniques and popular machine learning algorithms (Support Vector Machines, Random Forests and Neural Networks). We assess quantitative performance in two classification tasks: binary (transient/non-transient) and eight-class classification. The best performing algorithm in both task is the Random Forest Classifier. It achieves an F1-score of 86.61%in the binary classification and 50.38%in the eight-class classification. For the latter, the class with the highest F1-score are non-transients (87.12%)and the lowest corresponds to flares (11.96%); for supernovae it achieves a value of 50.07%, close to the average across classes.