The importance of precision in humour classification

摘要: Humour classification is one of the most interesting and difficult tasks in text classification. subjective by nature, yet humans are able to promptly define their preferences. Nowadays people often search for humour as a relaxing proxy overcome stressful demanding situations, having little or no time contents such activities. Hence, we propose aid definition personal models that allow user access with more confidence on precision his preferences. In this paper focus Support Vector Machine (SVM) active learning strategy uses specific informative examples improve baseline performance. Experiments were carried out using widely available Jester jokes dataset, encouraging results proposed framework.