Increases in the influx of data run in tandem with improvements in technology. And this is true across all fields of science, especially when you’re dealing with something as vast as the Universe itself. Machine learning techniques can be particularly helpful in this context reducing man-hours in sifting through terabytes of data when trying to find something of interest.

The aim of this project was to analyse and detect possible candidates for pulsars from a dataset made available by Lyon at el. Pulsars are a special type of star which emit an extremely powerful beam of electromagnetic radiation in a rotating fashion. Finding them against a background of cosmic radiation is a challenge demanding extensive expertise during manual classification. More often than not a candidate source is something entirely different. This means that we are dealing with an imbalanced dataset and presents itself as an excellent opportunity for me to learn how to deal with such cases.