Ongoing research has led to an interest in the relationship between gender and humour styles. This study aims to explore different machine learning models which are trained on data of a humour questionnaire. We find that the Support Vector Machine model is better than the others and predicts the gender with a higher accuracy score.
When comparing our models we adjust their parameters (hyperparameters) in order to generate the best possible predictions. We find the optimal values for each parameter in each model, which helps us to conclude which model is most appropriate for our question.
With these results, we could extend research into this relationship and perhaps develop other models which take into account additional factors when predicting the gender of a person.