Don't Take it Personally: Analyzing Gender and Age Differences in Ratings of Online Humor
This addresses the problem of subjectivity in computational humor detection for AI researchers, though it is incremental as it focuses on demographic analysis rather than a new detection method.
The study analyzed gender and age differences in humor and offense ratings, finding that women link humor and offense more strongly than men, give lower humor ratings and higher offense scores, and that this correlation increases with age.
Computational humor detection systems rarely model the subjectivity of humor responses, or consider alternative reactions to humor - namely offense. We analyzed a large dataset of humor and offense ratings by male and female annotators of different age groups. We find that women link these two concepts more strongly than men, and they tend to give lower humor ratings and higher offense scores. We also find that the correlation between humor and offense increases with age. Although there were no gender or age differences in humor detection, women and older annotators signalled that they did not understand joke texts more often than men. We discuss implications for computational humor detection and downstream tasks.